setFaviconError(false)}
/>
) : (
- GlobeAltIcon
+ GlobeLinesIcon
);
return (
diff --git a/apps/webapp/app/routes/account._index/route.tsx b/apps/webapp/app/routes/account._index/route.tsx
index 341141e638e..60d3975a1bd 100644
--- a/apps/webapp/app/routes/account._index/route.tsx
+++ b/apps/webapp/app/routes/account._index/route.tsx
@@ -1,7 +1,14 @@
import { getFormProps, getInputProps, useForm } from "@conform-to/react";
import { conformZodMessage, parseWithZod } from "@conform-to/zod";
-import { Form, type MetaFunction, useActionData } from "@remix-run/react";
-import { type ActionFunction, json } from "@remix-run/server-runtime";
+import { ComputerDesktopIcon, MoonIcon, SunIcon, SwatchIcon } from "@heroicons/react/20/solid";
+import {
+ Form,
+ type MetaFunction,
+ useActionData,
+ useFetcher,
+ useLoaderData,
+} from "@remix-run/react";
+import { type ActionFunction, json, type LoaderFunctionArgs } from "@remix-run/server-runtime";
import { z } from "zod";
import { UserProfilePhoto } from "~/components/UserProfilePhoto";
import {
@@ -10,6 +17,8 @@ import {
PageContainer,
} from "~/components/layout/AppLayout";
import { Button } from "~/components/primitives/Buttons";
+import { Select, SelectItem } from "~/components/primitives/Select";
+import { Slider } from "~/components/primitives/Slider";
import { FormError } from "~/components/primitives/FormError";
import { Header2 } from "~/components/primitives/Headers";
import { Input } from "~/components/primitives/Input";
@@ -21,10 +30,50 @@ import { prisma } from "~/db.server";
import { useUser } from "~/hooks/useUser";
import { redirectWithSuccessMessage } from "~/models/message.server";
import { updateUser } from "~/models/user.server";
-import { requireUserId } from "~/services/session.server";
+import {
+ updateContrastPreference,
+ updateThemePreference,
+} from "~/services/dashboardPreferences.server";
+import {
+ normalizeThemeContrast,
+ normalizeThemePreference,
+ type ThemePreference,
+} from "~/utils/themePreference";
+import { flag } from "~/v3/featureFlags.server";
+import { requireUser, requireUserId } from "~/services/session.server";
import { emailSchema, MAX_EMAIL_LENGTH } from "~/utils/emailValidation";
import { accountPath } from "~/utils/pathBuilder";
+const THEME_LABELS: Record = {
+ classic: "Classic",
+ system: "System preference",
+ dark: "Dark",
+ light: "Light",
+};
+
+function themeLabel(value: ThemePreference) {
+ return THEME_LABELS[value];
+}
+
+function themeIcon(value: ThemePreference) {
+ switch (value) {
+ case "classic":
+ return
+
+ );
+ case "light":
+ return ();
const lastSubmission = useActionData();
+ const themeFetcher = useFetcher();
+ const contrastFetcher = useFetcher();
+ const pendingTheme = themeFetcher.formData?.get("theme");
+ const pendingContrast = contrastFetcher.formData?.get("contrast");
+ const contrast =
+ typeof pendingContrast === "string"
+ ? normalizeThemeContrast(pendingContrast)
+ : normalizeThemeContrast(user.dashboardPreferences.contrast);
+ const theme: ThemePreference =
+ typeof pendingTheme === "string"
+ ? normalizeThemePreference(pendingTheme)
+ : normalizeThemePreference(user.dashboardPreferences.theme);
const [form, { name, email, marketingEmails }] = useForm({
id: "account",
@@ -207,6 +297,69 @@ export default function Page() {
+ {showThemeSwitcher && (
+ <>
+
+ Appearance
+
+
+ Interface theme
+
+ value={theme}
+ setValue={(value) =>
+ themeFetcher.submit(
+ { action: "update-theme", theme: value },
+ { method: "post" }
+ )
+ }
+ variant="secondary/small"
+ dropdownIcon
+ items={["classic", "system", "dark", "light"]}
+ text={(value) => (
+
+ {themeIcon(value)}
+ {themeLabel(value)}
+
+ )}
+ className="w-44"
+ >
+ {(items) =>
+ items.map((item) => (
+
+ {themeLabel(item)}
+
+ ))
+ }
+
+
+ {theme !== "classic" && (
+
+ Contrast
+ {
+ // Live preview before the preference persists
+ document.documentElement.style.setProperty(
+ "--theme-contrast",
+ String((values[0] ?? 0) / 100)
+ );
+ }}
+ onValueCommit={(values) =>
+ contrastFetcher.submit(
+ { action: "update-contrast", contrast: String(values[0] ?? 0) },
+ { method: "post" }
+ )
+ }
+ />
+
+ )}
+
+ )}
diff --git a/apps/webapp/app/routes/invites.tsx b/apps/webapp/app/routes/invites.tsx
index aa3b7809ca7..68522cd14d9 100644
--- a/apps/webapp/app/routes/invites.tsx
+++ b/apps/webapp/app/routes/invites.tsx
@@ -141,7 +141,7 @@ export default function Page() {
}
- className="mb-0 text-sky-500"
+ className="mb-0 text-sky-500 system:text-text-bright"
title={simplur`You have ${invites.length} new invitation[|s]`}
/>
{form.errors}
diff --git a/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.github.tsx b/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.github.tsx
index 1842edf347c..161f4281107 100644
--- a/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.github.tsx
+++ b/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.github.tsx
@@ -34,7 +34,7 @@ import { Label } from "~/components/primitives/Label";
import { Paragraph } from "~/components/primitives/Paragraph";
import { PermissionLink } from "~/components/primitives/PermissionLink";
import { Select, SelectItem } from "~/components/primitives/Select";
-import { SpinnerWhite } from "~/components/primitives/Spinner";
+import { Spinner } from "~/components/primitives/Spinner";
import { Switch } from "~/components/primitives/Switch";
import { TextLink } from "~/components/primitives/TextLink";
import { InfoIconTooltip } from "~/components/primitives/Tooltip";
@@ -596,7 +596,7 @@ export function ConnectGitHubRepoModal({
name="action"
value="connect-repo"
variant="primary/medium"
- LeadingIcon={isConnectRepositoryLoading ? SpinnerWhite : undefined}
+ LeadingIcon={isConnectRepositoryLoading ? Spinner : undefined}
leadingIconClassName="text-white"
disabled={isConnectRepositoryLoading}
>
@@ -907,7 +907,7 @@ export function ConnectedGitHubRepoForm({
? undefined
: "You don't have permission to manage the GitHub integration"
}
- LeadingIcon={isGitSettingsLoading ? SpinnerWhite : undefined}
+ LeadingIcon={isGitSettingsLoading ? Spinner : undefined}
>
Save
@@ -955,7 +955,7 @@ export function GitHubSettingsPanel({
if (fetcher.state === "loading" && !data) {
return (
- Loading GitHub settings...
);
diff --git a/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.runs.$runParam.spans.$spanParam/route.tsx b/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.runs.$runParam.spans.$spanParam/route.tsx
index 3527efdbf47..bdef2520eae 100644
--- a/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.runs.$runParam.spans.$spanParam/route.tsx
+++ b/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.runs.$runParam.spans.$spanParam/route.tsx
@@ -1,3 +1,4 @@
+import { GlobeLinesIcon } from "~/assets/icons/GlobeLinesIcon";
import {
ArrowPathIcon,
BookOpenIcon,
@@ -8,7 +9,6 @@ import {
CloudArrowDownIcon,
EnvelopeIcon,
ExclamationTriangleIcon,
- GlobeAltIcon,
KeyIcon,
QueueListIcon,
SignalIcon,
@@ -429,7 +429,9 @@ function RunBody({
@@ -720,7 +722,7 @@ function RunBody({
- Scope
@@ -1467,7 +1469,7 @@ function RunError({ error }: { error: TaskRunError }) {
{name}
{enhancedError.message && (
-
+
{enhancedError.message}
diff --git a/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.vercel.tsx b/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.vercel.tsx
index 7a5f61a48dc..ee84e52bfb7 100644
--- a/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.vercel.tsx
+++ b/apps/webapp/app/routes/resources.orgs.$organizationSlug.projects.$projectParam.env.$envParam.vercel.tsx
@@ -23,7 +23,7 @@ import { Label } from "~/components/primitives/Label";
import { Paragraph } from "~/components/primitives/Paragraph";
import { PermissionLink } from "~/components/primitives/PermissionLink";
import { Select, SelectItem } from "~/components/primitives/Select";
-import { SpinnerWhite } from "~/components/primitives/Spinner";
+import { Spinner } from "~/components/primitives/Spinner";
import {
redirectBackWithErrorMessage,
redirectWithErrorMessage,
@@ -562,7 +562,7 @@ function VercelConnectionPrompt({
}
LeadingIcon={
isLoadingProjects
- ? () =>
+ ? () =>
: () =>
}
>
@@ -993,7 +993,7 @@ function ConnectedVercelProjectForm({
LeadingIcon={
navigation.formData?.get("action") === "disable-auto-assign" &&
(navigation.state === "submitting" || navigation.state === "loading")
- ? SpinnerWhite
+ ? Spinner
: undefined
}
>
@@ -1022,7 +1022,7 @@ function ConnectedVercelProjectForm({
? undefined
: "You don't have permission to manage the Vercel integration"
}
- LeadingIcon={isConfigLoading ? SpinnerWhite : undefined}
+ LeadingIcon={isConfigLoading ? Spinner : undefined}
onClick={(event) => {
if (shouldPromptClearOnSave) {
event.preventDefault();
@@ -1125,7 +1125,7 @@ function VercelSettingsPanel({
Failed to load Vercel settings
-
+
There was an error loading the Vercel integration settings. Please refresh the page to
try again.
@@ -1138,7 +1138,7 @@ function VercelSettingsPanel({
if (fetcher.state === "loading" && !data) {
return (
- Loading Vercel settings...
);
diff --git a/apps/webapp/app/routes/resources.orgs.$organizationSlug.select-plan.tsx b/apps/webapp/app/routes/resources.orgs.$organizationSlug.select-plan.tsx
index 2c107c35b13..7bb6c1478e4 100644
--- a/apps/webapp/app/routes/resources.orgs.$organizationSlug.select-plan.tsx
+++ b/apps/webapp/app/routes/resources.orgs.$organizationSlug.select-plan.tsx
@@ -777,7 +777,7 @@ export function TierEnterprise() {
+
Contact us
}
diff --git a/apps/webapp/app/services/dashboardPreferences.server.ts b/apps/webapp/app/services/dashboardPreferences.server.ts
index a8b9149eff5..b25796b002a 100644
--- a/apps/webapp/app/services/dashboardPreferences.server.ts
+++ b/apps/webapp/app/services/dashboardPreferences.server.ts
@@ -25,8 +25,14 @@ export type SideMenuPreferences = z.infer;
import { type SideMenuSectionId } from "~/components/navigation/sideMenuTypes";
export type { SideMenuSectionId };
+import { ThemePreference } from "~/utils/themePreference";
+export { normalizeThemePreference, type ThemePreference } from "~/utils/themePreference";
+
const DashboardPreferences = z.object({
version: z.literal("1"),
+ theme: ThemePreference.optional(),
+ /** Interface contrast for the System themes, 0-100. */
+ contrast: z.number().int().min(0).max(100).optional(),
currentProjectId: z.string().optional(),
projects: z.record(
z.string(),
@@ -103,6 +109,67 @@ export async function updateCurrentProjectEnvironmentId({
});
}
+export async function updateThemePreference({
+ user,
+ theme,
+}: {
+ user: UserFromSession;
+ theme: ThemePreference;
+}) {
+ if (user.isImpersonating) {
+ return;
+ }
+
+ if (user.dashboardPreferences.theme === theme) {
+ return;
+ }
+
+ // Narrow jsonb_set write: a full-blob update from the session snapshot can
+ // race with other preference writes and drop unrelated fields.
+ return prisma.$executeRaw`
+ UPDATE "User"
+ SET "dashboardPreferences" = jsonb_set(
+ COALESCE(
+ "dashboardPreferences",
+ '{"version":"1","projects":{}}'::jsonb
+ ),
+ '{theme}',
+ to_jsonb(${theme}::text)
+ )
+ WHERE id = ${user.id}
+ `;
+}
+
+export async function updateContrastPreference({
+ user,
+ contrast,
+}: {
+ user: UserFromSession;
+ contrast: number;
+}) {
+ if (user.isImpersonating) {
+ return;
+ }
+
+ if (user.dashboardPreferences.contrast === contrast) {
+ return;
+ }
+
+ // Narrow jsonb_set write: see updateThemePreference.
+ return prisma.$executeRaw`
+ UPDATE "User"
+ SET "dashboardPreferences" = jsonb_set(
+ COALESCE(
+ "dashboardPreferences",
+ '{"version":"1","projects":{}}'::jsonb
+ ),
+ '{contrast}',
+ to_jsonb(${contrast}::int)
+ )
+ WHERE id = ${user.id}
+ `;
+}
+
export async function clearCurrentProject({ user }: { user: UserFromSession }) {
if (user.isImpersonating) {
return;
diff --git a/apps/webapp/app/tailwind.css b/apps/webapp/app/tailwind.css
index 2bc91d77323..f1acc4e3eb6 100644
--- a/apps/webapp/app/tailwind.css
+++ b/apps/webapp/app/tailwind.css
@@ -158,6 +158,7 @@
--color-surface-control: var(--color-charcoal-600);
--color-surface-control-hover: var(--color-charcoal-550);
--color-surface-control-active: var(--color-charcoal-500);
+ --color-input-bg: var(--color-charcoal-750);
/* Borders, from subtlest to most visible */
--color-grid-dimmed: var(--color-charcoal-750);
@@ -223,6 +224,84 @@
--color-callout-pricing-text: var(--color-indigo-300);
}
+/*
+ Theme overrides. A theme re-maps ONLY the semantic tokens above, never the raw
+ palette values. That invariant is what lets a future [data-contrast] overlay
+ re-map semantic tokens on top of whichever theme is active.
+*/
+
+/*
+ System preference unified accents (dark+light); Classic keeps the original
+ set. One shared value per accent token, >=3:1 against both dark cards and
+ white. Accents whose Classic default already clears both modes (blue-500,
+ indigo-500, pink-500, purple-500, red-500, violet-500, fuchsia-500,
+ rose-600) are not repeated here. Text-sized tokens (text-link, callout
+ text) stay per-mode: no color reaches 4.5:1 on both #1a1b1f and #ffffff.
+ Dark derives everything else (monochrome surfaces etc.) from Classic.
+*/
+:is([data-theme="dark"], [data-theme="light"]) {
+ /* Status */
+ --color-success: var(--color-mint-600);
+ --color-warning: var(--color-amber-600);
+ /* Env set: pink dev and green prod as in Classic, preview moves to blue
+ (yellow's unified mid-tone reads muddy brown). Staging stays warm orange
+ and is the one per-mode env color - see the light block. */
+ --color-prod: var(--color-mint-600);
+ --color-preview: var(--color-blue-500);
+
+ /* Icons */
+ --color-schedules: var(--color-yellow-700);
+ /* Matches the preview env color */
+ --color-previewBranches: var(--color-blue-500);
+ --color-metrics: var(--color-green-600);
+ --color-regions: var(--color-green-600);
+ --color-aiMetrics: var(--color-green-600);
+ --color-bulkActions: var(--color-emerald-600);
+ --color-concurrency: var(--color-amber-600);
+ --color-errors: var(--color-amber-600);
+ --color-apiKeys: var(--color-amber-600);
+
+ /* Queue charts read blue here; Classic keeps the queues purple */
+ --color-queues-chart: var(--color-blue-500);
+ --color-queues-chart-ref: var(--color-charcoal-500);
+
+ /* Callout accents (callout text/bg tints stay per-mode) */
+ --color-callout-docs: var(--color-blue-500);
+ --color-callout-pending: var(--color-blue-500);
+ --color-callout-pricing: var(--color-indigo-500);
+
+ /* Amber run statuses: three distinct steps, Classic's dark-to-light order
+ kept (waiting-for-deploy < pending-version < paused) */
+ --color-run-waiting-for-deploy: var(--color-amber-700);
+ --color-run-pending-version: #c76508;
+ --color-run-paused: var(--color-amber-600);
+ /* Nudged under 3:1's ceiling on white; stays lighter than system-failure */
+ --color-run-timed-out: #ed5f74;
+}
+
+/* System themes drop decorative icon accents to monochrome; Classic keeps the
+ colored icons. side-menu-active-icon is set in SideMenuItem for the active
+ nav item; system-mono-icon marks section-header icons (e.g. the limits page). */
+:is([data-theme="dark"], [data-theme="light"]) :is(.side-menu-active-icon, .system-mono-icon) {
+ color: var(--color-text-bright);
+}
+
+/* System themes: status/env labels follow the surrounding text color, only the
+ icon keeps its tint. Classic colors both. Set in EnvironmentLabel and the
+ status combo components. */
+:is([data-theme="dark"], [data-theme="light"]) .system-mono-label {
+ color: inherit;
+}
+
+/* Tinted status chips respond to the contrast control: a ring in the chip's
+ own text color fades in as contrast rises, so the soft tint keeps its
+ footprint but the chip gains definition. Transparent at contrast 0;
+ Classic never sets the variable. Marker set by the chip components. */
+:is([data-theme="dark"], [data-theme="light"]) .contrast-chip {
+ box-shadow: inset 0 0 0 1px
+ color-mix(in srgb, currentColor calc(var(--theme-contrast, 0) * 70%), transparent);
+}
+
/*
Code syntax palette - the trigger-dark highlight theme, shared by the shiki
theme (streamdown) and the prism theme (CodeBlock). Consumed from JS via
@@ -290,6 +369,14 @@
--color-editor-scrollbar-thumb-active: #3c4b62;
}
+/* Queue chart series colors - queues purple in Classic, overridden to blue in
+ the System themes; the grey reference series lightens per mode there.
+ Consumed from JS via var(), so declared static. */
+@theme static {
+ --color-queues-chart: var(--color-queues);
+ --color-queues-chart-ref: #4d525b;
+}
+
/*
Run-status chart colors. In-between shades are intentional - statuses within
a family (blues, roses, charcoals) are evenly spaced so chart series stay
@@ -333,8 +420,17 @@
@custom-variant lg-height (@media (max-height: 750px));
@custom-variant md-height (@media (max-height: 600px));
-/* dark: follows the app theme (data-theme on ), not the OS preference */
-@custom-variant dark (&:where([data-theme="dark"], [data-theme="dark"] *));
+/* dark: follows the app theme (data-theme on ), not the OS preference.
+ Classic and Dark are both dark-mode themes, so the variant matches both. */
+@custom-variant dark (&:where([data-theme="dark"], [data-theme="classic"], [data-theme="dark"] *, [data-theme="classic"] *));
+
+/* system: matches the System preference themes (Dark and Light) but never
+ Classic - for restyles that must leave Classic untouched. */
+@custom-variant system (&:where([data-theme="dark"], [data-theme="light"], [data-theme="dark"] *, [data-theme="light"] *));
+
+/* light: the Light theme only - for values that are fine on every dark theme
+ but illegible on white. Classic is never matched. */
+@custom-variant light (&:where([data-theme="light"], [data-theme="light"] *));
@utility focus-custom {
&:focus-visible {
@@ -489,7 +585,8 @@
--chart-5: 27 87% 67%;
}
- [data-theme="dark"] {
+ /* Classic and Dark share these HSL chart vars - they were the app-wide dark values. */
+ :is([data-theme="dark"], [data-theme="classic"]) {
--chart-1: 220 70% 50%;
--chart-2: 160 60% 45%;
--chart-3: 30 80% 55%;
@@ -497,6 +594,16 @@
--chart-5: 340 75% 55%;
}
+ /* System preference unified chart set (dark+light): Classic's hues, with
+ chart-2/3 darkened to clear 3:1 on white. Classic keeps the set above. */
+ :is([data-theme="dark"], [data-theme="light"]) {
+ --chart-1: 220 70% 50%;
+ --chart-2: 160 60% 40%;
+ --chart-3: 30 80% 47%;
+ --chart-4: 280 65% 60%;
+ --chart-5: 340 75% 55%;
+ }
+
/* Override react-grid-layout placeholder color (default is red) */
.react-grid-item.react-grid-placeholder {
background: rgb(99 102 241) !important; /* indigo-500 */
@@ -639,3 +746,236 @@
@apply leading-relaxed;
}
}
+
+/*
+ Light theme. Overrides the themable variables only; raw palettes stay put.
+ Code/editor values come from the trigger.light VS Code theme.
+*/
+[data-theme="light"] {
+ --color-secondary: #ffffff;
+ --color-input-bg: #ffffff;
+
+ /* shadcn vars consumed by charts/streamdown (tooltip cursor fill etc.) */
+ --background: #ffffff;
+ --foreground: #1a1b1f;
+ --muted: #eceef1;
+ --muted-foreground: #5f6570;
+ --border: #e2e4e9;
+ --sidebar: #f6f7f8;
+ --primary-foreground: #ffffff;
+
+ /* Text */
+ --color-primary: var(--color-apple-600);
+ --color-tertiary: #eef0f3;
+ --color-text-link: var(--color-lavender-600);
+ --color-text-faint: var(--color-charcoal-400);
+ --color-text-dimmed: var(--color-charcoal-500);
+ --color-text-bright: var(--color-charcoal-800);
+
+ /* Surfaces */
+ --color-background-deep: #f1f2f4;
+ --color-background-dimmed: #fbfbfc;
+ --color-background-bright: #ffffff;
+ --color-background-hover: #f2f3f5;
+ --color-background-raised: #e9eaee;
+ --color-surface-control: #dcdee3;
+ --color-surface-control-hover: #cfd2d9;
+ --color-surface-control-active: #b8bcc6;
+
+ /* Borders */
+ --color-grid-dimmed: #eceef1;
+ --color-grid-bright: #e2e4e9;
+ --color-border-bright: #d2d5db;
+ --color-border-brighter: #b9bdc7;
+ --color-border-brightest: #9ba1ad;
+
+ /* Status/env/icon accents live in the unified dark+light block above;
+ dev keeps its Classic pink-500, which passes on white too. */
+
+ /* Staging is the one per-mode env color: warm orange has no clean unified
+ mid-tone, so dark keeps Classic's bright orange-400 and light deepens it */
+ --color-staging: var(--color-orange-600);
+
+ /* Grey chart reference series, light enough not to dominate on white */
+ --color-queues-chart-ref: #c3c7cf;
+
+ /* Neutral run statuses: soft light grays for sparkbars and charts */
+ --color-run-pending: #ccd0d6;
+ --color-run-delayed: #d3d6db;
+ --color-run-waiting-to-resume: #c5c9d0;
+ --color-run-canceled: #d8dbdf;
+ --color-run-expired: #dee0e4;
+
+ /* Monochrome icon gray - per-mode, not part of the unified accent set */
+ --color-customDashboards: var(--color-charcoal-500);
+
+ /* Callout text/bg - deep tints instead of the dark theme's pastels; text
+ can't unify across modes (see the unified accents block) */
+ --color-callout-warning-text: var(--color-yellow-800);
+ --color-callout-error-text: var(--color-rose-700);
+ --color-callout-success-text: var(--color-green-800);
+ --color-callout-docs-text: var(--color-blue-800);
+ --color-callout-pending-bg: var(--color-blue-100);
+ --color-callout-pending-text: var(--color-blue-800);
+ --color-callout-pricing-bg: var(--color-indigo-100);
+ --color-callout-pricing-text: var(--color-indigo-800);
+
+ /* Code syntax - trigger.light */
+ --color-code-background: #ffffff;
+ --color-code-foreground: #333333;
+ --color-code-line-number: #a8a8ad;
+ --color-code-plain: #2e2e4b;
+ --color-code-muted: #333333;
+ --color-code-comment: #767a81;
+ --color-code-keyword: #b114d3;
+ --color-code-storage: #b114d3;
+ --color-code-type: #b114d3;
+ --color-code-function: #6532f5;
+ --color-code-variable: #404040;
+ --color-code-constant: #1e1e1e;
+ --color-code-language: #b114d3;
+ --color-code-object-key: #222222;
+ --color-code-string: #262626;
+ --color-code-template-punctuation: #0879e2;
+ --color-code-number: #262626;
+ --color-code-builtin: #3080e0;
+ --color-code-attribute: #222222;
+ --color-code-escape: #222222;
+ --color-code-regexp: #dc3545;
+ --color-code-regexp-constant: #5f6570;
+ --color-code-invalid: #dc3545;
+ --color-code-deleted: #dc3545;
+ --color-code-jsx-text: #2e2e4b;
+
+ /* CodeMirror - trigger.light */
+ --color-editor-background: #ffffff;
+ --color-editor-foreground: #2e2e4b;
+ --color-editor-keyword: #b114d3;
+ --color-editor-name: #197c3c;
+ --color-editor-function: #6532f5;
+ --color-editor-constant: #3080e0;
+ --color-editor-type: #2980b9;
+ --color-editor-operator: #333333;
+ --color-editor-comment: #767a81;
+ --color-editor-heading: #2c3e50;
+ --color-editor-string: #262626;
+ --color-editor-invalid: #dc3545;
+ --color-editor-panel-background: #f4f4f6;
+ --color-editor-highlight-background: #f8f8fa;
+ --color-editor-tooltip-background: #ffffff;
+ --color-editor-selection: #d9dce3;
+ --color-editor-cursor: #2e2e4b;
+ --color-editor-search-match: #0879e226;
+ --color-editor-search-match-outline: #0879e2;
+ --color-editor-search-match-selected: #0879e240;
+ --color-editor-selection-match: #e8e8ed;
+ --color-editor-matching-bracket: rgba(240, 241, 244, 0.9);
+ --color-editor-matching-bracket-outline: rgba(160, 166, 180, 0.5);
+ --color-editor-fold-placeholder: #555555;
+ --color-editor-scrollbar-track-active: #e8e9ec;
+ --color-editor-scrollbar-thumb: #c9ccd4;
+ --color-editor-scrollbar-thumb-active: #aeb3be;
+}
+
+/* Streamdown's muted surface has no semantic token (charcoal-775); theme it here */
+[data-theme="light"] .streamdown-container {
+ --muted: #eceef1;
+}
+
+/* The timeline label shadow is a Classic legibility aid; the System themes
+ drop it entirely */
+:is([data-theme="dark"], [data-theme="light"]) .text-shadow-custom {
+ text-shadow: none;
+}
+
+/* Neutral timeline points: invert to a light dot with a gray ring */
+[data-theme="light"] .timeline-point.bg-surface-control-active {
+ border-color: var(--color-surface-control-active);
+ background-color: var(--color-background-bright);
+}
+
+/* Run timeline bars: no fade gradient on light */
+[data-theme="light"] .timeline-span {
+ background-image: none;
+}
+/* On saturated bars the duration label keeps the dark-theme treatment,
+ but only when the bar is wide enough to contain the label — on narrow
+ bars the sticky label overflows onto the page background, where the
+ default dark-on-light text is correct. */
+[data-theme="light"] .timeline-span.bg-success,
+[data-theme="light"] .timeline-span.bg-error,
+[data-theme="light"] .timeline-span.bg-blue-500 {
+ container-type: inline-size;
+}
+@container (min-width: 3.5rem) {
+ [data-theme="light"] .timeline-span.bg-success .text-shadow-custom,
+ [data-theme="light"] .timeline-span.bg-error .text-shadow-custom,
+ [data-theme="light"] .timeline-span.bg-blue-500 .text-shadow-custom {
+ color: #ffffff;
+ }
+}
+
+/* The table row-hover menu wraps the always-visible "Suggest a region" button
+ in a ring container; on light that ring doubles up with the button's own
+ border, so drop it here. */
+[data-theme="light"] .suggest-region-cell > div > div {
+ box-shadow: none;
+}
+
+/*
+ Interface contrast (System themes only). --theme-contrast is 0..1, set on
+ from the dashboard preference. It pulls the whole monochrome scale
+ apart: text mixes toward the mode's foreground, backgrounds toward its
+ depth, borders step up. Accents are untouched. srgb mixing keeps contrast 0
+ byte-identical to the base values, and Classic never reads the variable.
+ These blocks must stay below the [data-theme="light"] theme block so they
+ win the cascade.
+*/
+[data-theme="dark"] {
+ /* Text brightens toward white */
+ --color-text-bright: color-mix(in srgb, var(--color-charcoal-200), #fff calc(var(--theme-contrast, 0) * 100%));
+ --color-text-dimmed: color-mix(in srgb, var(--color-charcoal-400), #fff calc(var(--theme-contrast, 0) * 60%));
+ --color-text-faint: color-mix(in srgb, var(--color-charcoal-500), #fff calc(var(--theme-contrast, 0) * 45%));
+
+ /* Backgrounds deepen toward black, keeping their relative order */
+ --color-background-deep: color-mix(in srgb, var(--color-charcoal-900), #000 calc(var(--theme-contrast, 0) * 60%));
+ --color-background-dimmed: color-mix(in srgb, var(--color-charcoal-850), #000 calc(var(--theme-contrast, 0) * 55%));
+ --color-background-bright: color-mix(in srgb, var(--color-charcoal-800), #000 calc(var(--theme-contrast, 0) * 45%));
+ --color-background-hover: color-mix(in srgb, var(--color-charcoal-750), #000 calc(var(--theme-contrast, 0) * 35%));
+ --color-background-raised: color-mix(in srgb, var(--color-charcoal-700), #000 calc(var(--theme-contrast, 0) * 25%));
+ --color-input-bg: color-mix(in srgb, var(--color-charcoal-750), #000 calc(var(--theme-contrast, 0) * 35%));
+
+ /* Controls and borders step up toward white */
+ --color-surface-control: color-mix(in srgb, var(--color-charcoal-600), #fff calc(var(--theme-contrast, 0) * 12%));
+ --color-surface-control-hover: color-mix(in srgb, var(--color-charcoal-550), #fff calc(var(--theme-contrast, 0) * 14%));
+ --color-surface-control-active: color-mix(in srgb, var(--color-charcoal-500), #fff calc(var(--theme-contrast, 0) * 16%));
+ --color-grid-dimmed: color-mix(in srgb, var(--color-charcoal-750), #fff calc(var(--theme-contrast, 0) * 18%));
+ --color-grid-bright: color-mix(in srgb, var(--color-charcoal-700), #fff calc(var(--theme-contrast, 0) * 20%));
+ --color-border-bright: color-mix(in srgb, var(--color-charcoal-600), #fff calc(var(--theme-contrast, 0) * 24%));
+ --color-border-brighter: color-mix(in srgb, var(--color-charcoal-550), #fff calc(var(--theme-contrast, 0) * 27%));
+ --color-border-brightest: color-mix(in srgb, var(--color-charcoal-500), #fff calc(var(--theme-contrast, 0) * 30%));
+}
+
+[data-theme="light"] {
+ /* Text deepens toward black */
+ --color-text-bright: color-mix(in srgb, #1a1b1f, #000 calc(var(--theme-contrast, 0) * 100%));
+ --color-text-dimmed: color-mix(in srgb, var(--color-charcoal-500), #000 calc(var(--theme-contrast, 0) * 85%));
+ --color-text-faint: color-mix(in srgb, var(--color-charcoal-400), #000 calc(var(--theme-contrast, 0) * 70%));
+
+ /* On white the foregrounds carry the contrast: cards stay white while the
+ page-behind surfaces darken a touch so panels separate */
+ --color-background-deep: color-mix(in srgb, #f1f2f4, #000 calc(var(--theme-contrast, 0) * 10%));
+ --color-background-hover: color-mix(in srgb, #f2f3f5, #000 calc(var(--theme-contrast, 0) * 8%));
+ --color-background-raised: color-mix(in srgb, #e9eaee, #000 calc(var(--theme-contrast, 0) * 10%));
+
+ /* Controls and borders push hard toward black - this is where light-mode
+ contrast is actually visible */
+ --color-surface-control: color-mix(in srgb, #dcdee3, #000 calc(var(--theme-contrast, 0) * 25%));
+ --color-surface-control-hover: color-mix(in srgb, #cfd2d9, #000 calc(var(--theme-contrast, 0) * 28%));
+ --color-surface-control-active: color-mix(in srgb, #b8bcc6, #000 calc(var(--theme-contrast, 0) * 32%));
+ --color-grid-dimmed: color-mix(in srgb, #eceef1, #000 calc(var(--theme-contrast, 0) * 28%));
+ --color-grid-bright: color-mix(in srgb, #e2e4e9, #000 calc(var(--theme-contrast, 0) * 32%));
+ --color-border-bright: color-mix(in srgb, #d2d5db, #000 calc(var(--theme-contrast, 0) * 38%));
+ --color-border-brighter: color-mix(in srgb, #b9bdc7, #000 calc(var(--theme-contrast, 0) * 42%));
+ --color-border-brightest: color-mix(in srgb, #9ba1ad, #000 calc(var(--theme-contrast, 0) * 46%));
+}
diff --git a/apps/webapp/app/utils/themePreference.ts b/apps/webapp/app/utils/themePreference.ts
new file mode 100644
index 00000000000..2b6408c2abc
--- /dev/null
+++ b/apps/webapp/app/utils/themePreference.ts
@@ -0,0 +1,25 @@
+import { z } from "zod";
+
+// Shared between server (dashboard preferences) and client (theme UI, system
+// theme sync) - must stay free of server-only imports.
+export const ThemePreference = z.enum(["classic", "system", "dark", "light"]);
+export type ThemePreference = z.infer;
+
+/** Coerce any stored/legacy value into a valid preference. Missing or unknown
+ * values fall back to `dark` - the new dark theme is the default (pinned, not
+ * system-resolved, so nobody gets surprised by light mode). */
+export function normalizeThemePreference(value: unknown): ThemePreference {
+ const result = ThemePreference.safeParse(value);
+ return result.success ? result.data : "dark";
+}
+
+/** The default dark theme ships with a slight contrast bump. */
+export const DEFAULT_THEME_CONTRAST = 50;
+
+/** Interface contrast for the System themes, 0 to 100. Missing or invalid
+ * values fall back to the default bump. */
+export function normalizeThemeContrast(value: unknown): number {
+ const num = typeof value === "string" ? Number(value) : value;
+ if (typeof num !== "number" || !Number.isFinite(num)) return DEFAULT_THEME_CONTRAST;
+ return Math.min(100, Math.max(0, Math.round(num)));
+}
diff --git a/apps/webapp/app/v3/featureFlags.ts b/apps/webapp/app/v3/featureFlags.ts
index eefb5d665ef..1b7694b220b 100644
--- a/apps/webapp/app/v3/featureFlags.ts
+++ b/apps/webapp/app/v3/featureFlags.ts
@@ -10,6 +10,7 @@ export const FEATURE_FLAG = {
hasComputeAccess: "hasComputeAccess",
hasPrivateConnections: "hasPrivateConnections",
hasSso: "hasSso",
+ hasThemeSwitcher: "hasThemeSwitcher",
mollifierEnabled: "mollifierEnabled",
workerQueueScheduledSplitEnabled: "workerQueueScheduledSplitEnabled",
realtimeBackend: "realtimeBackend",
@@ -37,6 +38,8 @@ export const FeatureFlagCatalog = {
[FEATURE_FLAG.hasComputeAccess]: z.coerce.boolean(),
[FEATURE_FLAG.hasPrivateConnections]: z.coerce.boolean(),
[FEATURE_FLAG.hasSso]: z.coerce.boolean(),
+ // Gates the Interface theme setting in /account. Off by default.
+ [FEATURE_FLAG.hasThemeSwitcher]: z.coerce.boolean(),
[FEATURE_FLAG.mollifierEnabled]: z.coerce.boolean(),
[FEATURE_FLAG.workerQueueScheduledSplitEnabled]: z.coerce.boolean(),
// Which backend serves the realtime run feed. Controllable
diff --git a/apps/webapp/test/themePreference.test.ts b/apps/webapp/test/themePreference.test.ts
new file mode 100644
index 00000000000..b6656fefb38
--- /dev/null
+++ b/apps/webapp/test/themePreference.test.ts
@@ -0,0 +1,43 @@
+import { describe, expect, it } from "vitest";
+import { getDashboardPreferences } from "~/services/dashboardPreferences.server";
+import { normalizeThemePreference, type ThemePreference } from "~/utils/themePreference";
+
+const VALID_THEMES: ThemePreference[] = ["classic", "system", "dark", "light"];
+
+describe("normalizeThemePreference", () => {
+ it("returns each valid value unchanged", () => {
+ for (const theme of VALID_THEMES) {
+ expect(normalizeThemePreference(theme)).toBe(theme);
+ }
+ });
+
+ it("falls back to dark for legacy/unknown values", () => {
+ expect(normalizeThemePreference("solarized")).toBe("dark");
+ expect(normalizeThemePreference("")).toBe("dark");
+ expect(normalizeThemePreference(42)).toBe("dark");
+ expect(normalizeThemePreference(null)).toBe("dark");
+ });
+
+ it("falls back to dark for undefined", () => {
+ expect(normalizeThemePreference(undefined)).toBe("dark");
+ });
+});
+
+describe("DashboardPreferences theme schema", () => {
+ it("accepts all four theme values", () => {
+ for (const theme of VALID_THEMES) {
+ const result = getDashboardPreferences({ version: "1", projects: {}, theme });
+ expect(result.theme).toBe(theme);
+ }
+ });
+
+ it("accepts preferences without a theme", () => {
+ const result = getDashboardPreferences({ version: "1", projects: {} });
+ expect(result.theme).toBeUndefined();
+ });
+
+ it("rejects an invalid theme by falling back to defaults", () => {
+ const result = getDashboardPreferences({ version: "1", projects: {}, theme: "neon" });
+ expect(result.theme).toBeUndefined();
+ });
+});
diff --git a/internal-packages/llm-model-catalog/src/defaultPrices.ts b/internal-packages/llm-model-catalog/src/defaultPrices.ts
index 982b6b2ec15..fb347c2bef6 100644
--- a/internal-packages/llm-model-catalog/src/defaultPrices.ts
+++ b/internal-packages/llm-model-catalog/src/defaultPrices.ts
@@ -6,3106 +6,3091 @@ import type { DefaultModelDefinition } from "./types.js";
export const defaultModelPrices: DefaultModelDefinition[] = [
{
- modelName: "gpt-4o",
- matchPattern: "(?i)^(openai/)?(gpt-4o)$",
- startDate: "2024-05-13T23:15:07.670Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000025,
- input_cached_tokens: 0.00000125,
- input_cache_read: 0.00000125,
- output: 0.00001,
- },
- },
- ],
- },
- {
- modelName: "gpt-4o-2024-05-13",
- matchPattern: "(?i)^(openai/)?(gpt-4o-2024-05-13)$",
- startDate: "2024-05-13T23:15:07.670Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000005,
- output: 0.000015,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-1106-preview",
- matchPattern: "(?i)^(openai/)?(gpt-4-1106-preview)$",
- startDate: "2024-04-23T10:37:17.092Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00001,
- output: 0.00003,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-turbo-vision",
- matchPattern: "(?i)^(openai/)?(gpt-4(-\\d{4})?-vision-preview)$",
- startDate: "2024-01-24T10:19:21.693Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00001,
- output: 0.00003,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-32k",
- matchPattern: "(?i)^(openai/)?(gpt-4-32k)$",
- startDate: "2024-01-24T10:19:21.693Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00006,
- output: 0.00012,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-32k-0613",
- matchPattern: "(?i)^(openai/)?(gpt-4-32k-0613)$",
- startDate: "2024-01-24T10:19:21.693Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00006,
- output: 0.00012,
- },
- },
- ],
- },
- {
- modelName: "gpt-3.5-turbo-1106",
- matchPattern: "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-1106)$",
- startDate: "2024-01-24T10:19:21.693Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000001,
- output: 0.000002,
- },
- },
- ],
- },
- {
- modelName: "gpt-3.5-turbo-0613",
- matchPattern: "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-0613)$",
- startDate: "2024-01-24T10:19:21.693Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000015,
- output: 0.000002,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-0613",
- matchPattern: "(?i)^(openai/)?(gpt-4-0613)$",
- startDate: "2024-01-24T10:19:21.693Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00003,
- output: 0.00006,
- },
- },
- ],
- },
- {
- modelName: "gpt-3.5-turbo-instruct",
- matchPattern: "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-instruct)$",
- startDate: "2024-01-24T10:19:21.693Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000015,
- output: 0.000002,
- },
- },
- ],
- },
- {
- modelName: "text-ada-001",
- matchPattern: "(?i)^(text-ada-001)$",
- startDate: "2024-01-24T18:18:50.861Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 0.000004,
- },
- },
- ],
- },
- {
- modelName: "text-babbage-001",
- matchPattern: "(?i)^(text-babbage-001)$",
- startDate: "2024-01-24T18:18:50.861Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "text-curie-001",
- matchPattern: "(?i)^(text-curie-001)$",
- startDate: "2024-01-24T18:18:50.861Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 0.00002,
- },
- },
- ],
- },
- {
- modelName: "text-davinci-001",
- matchPattern: "(?i)^(text-davinci-001)$",
- startDate: "2024-01-24T18:18:50.861Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 0.00002,
- },
- },
- ],
- },
- {
- modelName: "text-davinci-002",
- matchPattern: "(?i)^(text-davinci-002)$",
- startDate: "2024-01-24T18:18:50.861Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 0.00002,
- },
- },
- ],
- },
- {
- modelName: "text-davinci-003",
- matchPattern: "(?i)^(text-davinci-003)$",
- startDate: "2024-01-24T18:18:50.861Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 0.00002,
- },
- },
- ],
- },
- {
- modelName: "text-embedding-ada-002-v2",
- matchPattern: "(?i)^(text-embedding-ada-002-v2)$",
- startDate: "2024-01-24T18:18:50.861Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 1e-7,
- },
- },
- ],
- },
- {
- modelName: "text-embedding-ada-002",
- matchPattern: "(?i)^(text-embedding-ada-002)$",
- startDate: "2024-01-24T18:18:50.861Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 1e-7,
- },
- },
- ],
- },
- {
- modelName: "gpt-3.5-turbo-16k-0613",
- matchPattern: "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-16k-0613)$",
- startDate: "2024-02-03T17:29:57.350Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- output: 0.000004,
- },
- },
- ],
- },
- {
- modelName: "gpt-3.5-turbo-0301",
- matchPattern: "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-0301)$",
- startDate: "2024-01-24T10:19:21.693Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000002,
- output: 0.000002,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-32k-0314",
- matchPattern: "(?i)^(openai/)?(gpt-4-32k-0314)$",
- startDate: "2024-01-24T10:19:21.693Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00006,
- output: 0.00012,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-0314",
- matchPattern: "(?i)^(openai/)?(gpt-4-0314)$",
- startDate: "2024-01-24T10:19:21.693Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00003,
- output: 0.00006,
- },
- },
- ],
- },
- {
- modelName: "gpt-4",
- matchPattern: "(?i)^(openai/)?(gpt-4)$",
- startDate: "2024-01-24T10:19:21.693Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00003,
- output: 0.00006,
- },
- },
- ],
- },
- {
- modelName: "claude-instant-1.2",
- matchPattern: "(?i)^(anthropic/)?(claude-instant-1.2)$",
- startDate: "2024-01-30T15:44:13.447Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00000163,
- output: 0.00000551,
- },
- },
- ],
- },
- {
- modelName: "claude-2.0",
- matchPattern: "(?i)^(anthropic/)?(claude-2.0)$",
- startDate: "2024-01-30T15:44:13.447Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000008,
- output: 0.000024,
- },
- },
- ],
- },
- {
- modelName: "claude-2.1",
- matchPattern: "(?i)^(anthropic/)?(claude-2.1)$",
- startDate: "2024-01-30T15:44:13.447Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000008,
- output: 0.000024,
- },
- },
- ],
- },
- {
- modelName: "claude-1.3",
- matchPattern: "(?i)^(anthropic/)?(claude-1.3)$",
- startDate: "2024-01-30T15:44:13.447Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000008,
- output: 0.000024,
- },
- },
- ],
- },
- {
- modelName: "claude-1.2",
- matchPattern: "(?i)^(anthropic/)?(claude-1.2)$",
- startDate: "2024-01-30T15:44:13.447Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000008,
- output: 0.000024,
- },
- },
- ],
- },
- {
- modelName: "claude-1.1",
- matchPattern: "(?i)^(anthropic/)?(claude-1.1)$",
- startDate: "2024-01-30T15:44:13.447Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000008,
- output: 0.000024,
- },
- },
- ],
- },
- {
- modelName: "claude-instant-1",
- matchPattern: "(?i)^(anthropic/)?(claude-instant-1)$",
- startDate: "2024-01-30T15:44:13.447Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00000163,
- output: 0.00000551,
- },
- },
- ],
- },
- {
- modelName: "babbage-002",
- matchPattern: "(?i)^(babbage-002)$",
- startDate: "2024-01-26T17:35:21.129Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 4e-7,
- output: 0.0000016,
- },
- },
- ],
- },
- {
- modelName: "davinci-002",
- matchPattern: "(?i)^(davinci-002)$",
- startDate: "2024-01-26T17:35:21.129Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000006,
- output: 0.000012,
- },
- },
- ],
- },
- {
- modelName: "text-embedding-3-small",
- matchPattern: "(?i)^(text-embedding-3-small)$",
- startDate: "2024-01-26T17:35:21.129Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 2e-8,
- },
- },
- ],
- },
- {
- modelName: "text-embedding-3-large",
- matchPattern: "(?i)^(text-embedding-3-large)$",
- startDate: "2024-01-26T17:35:21.129Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 1.3e-7,
- },
- },
- ],
- },
- {
- modelName: "gpt-3.5-turbo-0125",
- matchPattern: "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-0125)$",
- startDate: "2024-01-26T17:35:21.129Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 5e-7,
- output: 0.0000015,
- },
- },
- ],
- },
- {
- modelName: "gpt-3.5-turbo",
- matchPattern: "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo)$",
- startDate: "2024-02-13T12:00:37.424Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 5e-7,
- output: 0.0000015,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-0125-preview",
- matchPattern: "(?i)^(openai/)?(gpt-4-0125-preview)$",
- startDate: "2024-01-26T17:35:21.129Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00001,
- output: 0.00003,
- },
- },
- ],
- },
- {
- modelName: "ft:gpt-3.5-turbo-1106",
- matchPattern: "(?i)^(ft:)(gpt-3.5-turbo-1106:)(.+)(:)(.*)(:)(.+)$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- output: 0.000006,
- },
- },
- ],
- },
- {
- modelName: "ft:gpt-3.5-turbo-0613",
- matchPattern: "(?i)^(ft:)(gpt-3.5-turbo-0613:)(.+)(:)(.*)(:)(.+)$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000012,
- output: 0.000016,
- },
- },
- ],
- },
- {
- modelName: "ft:davinci-002",
- matchPattern: "(?i)^(ft:)(davinci-002:)(.+)(:)(.*)(:)(.+)$$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000012,
- output: 0.000012,
- },
- },
- ],
- },
- {
- modelName: "ft:babbage-002",
- matchPattern: "(?i)^(ft:)(babbage-002:)(.+)(:)(.*)(:)(.+)$$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000016,
- output: 0.0000016,
- },
- },
- ],
- },
- {
- modelName: "chat-bison",
- matchPattern: "(?i)^(chat-bison)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- output: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "codechat-bison-32k",
- matchPattern: "(?i)^(codechat-bison-32k)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- output: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "codechat-bison",
- matchPattern: "(?i)^(codechat-bison)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- output: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "text-bison-32k",
- matchPattern: "(?i)^(text-bison-32k)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- output: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "chat-bison-32k",
- matchPattern: "(?i)^(chat-bison-32k)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- output: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "text-unicorn",
- matchPattern: "(?i)^(text-unicorn)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000025,
- output: 0.0000075,
- },
- },
- ],
- },
- {
- modelName: "text-bison",
- matchPattern: "(?i)^(text-bison)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- output: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "textembedding-gecko",
- matchPattern: "(?i)^(textembedding-gecko)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 1e-7,
- },
- },
- ],
- },
- {
- modelName: "textembedding-gecko-multilingual",
- matchPattern: "(?i)^(textembedding-gecko-multilingual)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- total: 1e-7,
- },
- },
- ],
- },
- {
- modelName: "code-gecko",
- matchPattern: "(?i)^(code-gecko)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- output: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "code-bison",
- matchPattern: "(?i)^(code-bison)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- output: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "code-bison-32k",
- matchPattern: "(?i)^(code-bison-32k)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-01-31T13:25:02.141Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- output: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "gpt-3.5-turbo-16k",
- matchPattern: "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-16k)$",
- startDate: "2024-02-13T12:00:37.424Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 5e-7,
- output: 0.0000015,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-turbo-preview",
- matchPattern: "(?i)^(openai/)?(gpt-4-turbo-preview)$",
- startDate: "2024-02-15T21:21:50.947Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00001,
- output: 0.00003,
- },
- },
- ],
- },
- {
- modelName: "claude-3-opus-20240229",
- matchPattern:
- "(?i)^(anthropic/)?(claude-3-opus-20240229|anthropic\\.claude-3-opus-20240229-v1:0|claude-3-opus@20240229)$",
- startDate: "2024-03-07T17:55:38.139Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000015,
- output: 0.000075,
- },
- },
- ],
- },
- {
- modelName: "claude-3-sonnet-20240229",
- matchPattern:
- "(?i)^(anthropic/)?(claude-3-sonnet-20240229|anthropic\\.claude-3-sonnet-20240229-v1:0|claude-3-sonnet@20240229)$",
- startDate: "2024-03-07T17:55:38.139Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- input_tokens: 0.000003,
- output: 0.000015,
- output_tokens: 0.000015,
- cache_creation_input_tokens: 0.00000375,
- input_cache_creation: 0.00000375,
- input_cache_creation_5m: 0.00000375,
- input_cache_creation_1h: 0.000006,
- cache_read_input_tokens: 3e-7,
- input_cache_read: 3e-7,
- },
- },
- ],
- },
- {
- modelName: "claude-3-haiku-20240307",
- matchPattern:
- "(?i)^(anthropic/)?(claude-3-haiku-20240307|anthropic\\.claude-3-haiku-20240307-v1:0|claude-3-haiku@20240307)$",
- startDate: "2024-03-14T09:41:18.736Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- output: 0.00000125,
- },
- },
- ],
- },
- {
- modelName: "gemini-1.0-pro-latest",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-1.0-pro-latest)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-04-11T10:27:46.517Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- output: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "gemini-1.0-pro",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-1.0-pro)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-04-11T10:27:46.517Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 1.25e-7,
- output: 3.75e-7,
- },
- },
- ],
- },
- {
- modelName: "gemini-1.0-pro-001",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-1.0-pro-001)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-04-11T10:27:46.517Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 1.25e-7,
- output: 3.75e-7,
- },
- },
- ],
- },
- {
- modelName: "gemini-pro",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-pro)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-04-11T10:27:46.517Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 1.25e-7,
- output: 3.75e-7,
- },
- },
- ],
- },
- {
- modelName: "gemini-1.5-pro-latest",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-1.5-pro-latest)(@[a-zA-Z0-9]+)?$",
- startDate: "2024-04-11T10:27:46.517Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000025,
- output: 0.0000075,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-turbo-2024-04-09",
- matchPattern: "(?i)^(openai/)?(gpt-4-turbo-2024-04-09)$",
- startDate: "2024-04-23T10:37:17.092Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00001,
- output: 0.00003,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-turbo",
- matchPattern: "(?i)^(openai/)?(gpt-4-turbo)$",
- startDate: "2024-04-11T21:13:44.989Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00001,
- output: 0.00003,
- },
- },
- ],
- },
- {
- modelName: "gpt-4-preview",
- matchPattern: "(?i)^(openai/)?(gpt-4-preview)$",
- startDate: "2024-04-23T10:37:17.092Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00001,
- output: 0.00003,
- },
- },
- ],
- },
- {
- modelName: "claude-3-5-sonnet-20240620",
- matchPattern:
- "(?i)^(anthropic/)?(claude-3-5-sonnet-20240620|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-3-5-sonnet-20240620-v1:0|claude-3-5-sonnet@20240620)$",
- startDate: "2024-06-25T11:47:24.475Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- input_tokens: 0.000003,
- output: 0.000015,
- output_tokens: 0.000015,
- cache_creation_input_tokens: 0.00000375,
- input_cache_creation: 0.00000375,
- input_cache_creation_5m: 0.00000375,
- input_cache_creation_1h: 0.000006,
- cache_read_input_tokens: 3e-7,
- input_cache_read: 3e-7,
- },
- },
- ],
- },
- {
- modelName: "gpt-4o-mini",
- matchPattern: "(?i)^(openai/)?(gpt-4o-mini)$",
- startDate: "2024-07-18T17:56:09.591Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 1.5e-7,
- output: 6e-7,
- input_cached_tokens: 7.5e-8,
- input_cache_read: 7.5e-8,
- },
- },
- ],
- },
- {
- modelName: "gpt-4o-mini-2024-07-18",
- matchPattern: "(?i)^(openai/)?(gpt-4o-mini-2024-07-18)$",
- startDate: "2024-07-18T17:56:09.591Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 1.5e-7,
- input_cached_tokens: 7.5e-8,
- input_cache_read: 7.5e-8,
- output: 6e-7,
- },
- },
- ],
- },
- {
- modelName: "gpt-4o-2024-08-06",
- matchPattern: "(?i)^(openai/)?(gpt-4o-2024-08-06)$",
- startDate: "2024-08-07T11:54:31.298Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000025,
- input_cached_tokens: 0.00000125,
- input_cache_read: 0.00000125,
- output: 0.00001,
- },
- },
- ],
- },
- {
- modelName: "o1-preview",
- matchPattern: "(?i)^(openai/)?(o1-preview)$",
- startDate: "2024-09-13T10:01:35.373Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000015,
- input_cached_tokens: 0.0000075,
- input_cache_read: 0.0000075,
- output: 0.00006,
- output_reasoning_tokens: 0.00006,
- output_reasoning: 0.00006,
- },
- },
- ],
- },
- {
- modelName: "o1-preview-2024-09-12",
- matchPattern: "(?i)^(openai/)?(o1-preview-2024-09-12)$",
- startDate: "2024-09-13T10:01:35.373Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000015,
- input_cached_tokens: 0.0000075,
- input_cache_read: 0.0000075,
- output: 0.00006,
- output_reasoning_tokens: 0.00006,
- output_reasoning: 0.00006,
- },
- },
- ],
- },
- {
- modelName: "o1-mini",
- matchPattern: "(?i)^(openai/)?(o1-mini)$",
- startDate: "2024-09-13T10:01:35.373Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000011,
- input_cached_tokens: 5.5e-7,
- input_cache_read: 5.5e-7,
- output: 0.0000044,
- output_reasoning_tokens: 0.0000044,
- output_reasoning: 0.0000044,
- },
- },
- ],
- },
- {
- modelName: "o1-mini-2024-09-12",
- matchPattern: "(?i)^(openai/)?(o1-mini-2024-09-12)$",
- startDate: "2024-09-13T10:01:35.373Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000011,
- input_cached_tokens: 5.5e-7,
- input_cache_read: 5.5e-7,
- output: 0.0000044,
- output_reasoning_tokens: 0.0000044,
- output_reasoning: 0.0000044,
- },
- },
- ],
- },
- {
- modelName: "claude-3.5-sonnet-20241022",
- matchPattern:
- "(?i)^(anthropic/)?(claude-3-5-sonnet-20241022|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-3-5-sonnet-20241022-v2:0|claude-3-5-sonnet-V2@20241022)$",
- startDate: "2024-10-22T18:48:01.676Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- input_tokens: 0.000003,
- output: 0.000015,
- output_tokens: 0.000015,
- cache_creation_input_tokens: 0.00000375,
- input_cache_creation: 0.00000375,
- input_cache_creation_5m: 0.00000375,
- input_cache_creation_1h: 0.000006,
- cache_read_input_tokens: 3e-7,
- input_cache_read: 3e-7,
- },
- },
- ],
- },
- {
- modelName: "claude-3.5-sonnet-latest",
- matchPattern: "(?i)^(anthropic/)?(claude-3-5-sonnet-latest)$",
- startDate: "2024-10-22T18:48:01.676Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- input_tokens: 0.000003,
- output: 0.000015,
- output_tokens: 0.000015,
- cache_creation_input_tokens: 0.00000375,
- input_cache_creation: 0.00000375,
- input_cache_creation_5m: 0.00000375,
- input_cache_creation_1h: 0.000006,
- cache_read_input_tokens: 3e-7,
- input_cache_read: 3e-7,
- },
- },
- ],
- },
- {
- modelName: "claude-3-5-haiku-20241022",
- matchPattern:
- "(?i)^(anthropic/)?(claude-3-5-haiku-20241022|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-3-5-haiku-20241022-v1:0|claude-3-5-haiku-V1@20241022)$",
- startDate: "2024-11-05T10:30:50.566Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 8e-7,
- input_tokens: 8e-7,
- output: 0.000004,
- output_tokens: 0.000004,
- cache_creation_input_tokens: 0.000001,
- input_cache_creation: 0.000001,
- input_cache_creation_5m: 0.000001,
- input_cache_creation_1h: 0.0000016,
- cache_read_input_tokens: 8e-8,
- input_cache_read: 8e-8,
- },
- },
- ],
- },
- {
- modelName: "claude-3.5-haiku-latest",
- matchPattern: "(?i)^(anthropic/)?(claude-3-5-haiku-latest)$",
- startDate: "2024-11-05T10:30:50.566Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 8e-7,
- input_tokens: 8e-7,
- output: 0.000004,
- output_tokens: 0.000004,
- cache_creation_input_tokens: 0.000001,
- input_cache_creation: 0.000001,
- input_cache_creation_5m: 0.000001,
- input_cache_creation_1h: 0.0000016,
- cache_read_input_tokens: 8e-8,
- input_cache_read: 8e-8,
- },
- },
- ],
- },
- {
- modelName: "chatgpt-4o-latest",
- matchPattern: "(?i)^(chatgpt-4o-latest)$",
- startDate: "2024-11-25T12:47:17.504Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000005,
- output: 0.000015,
- },
- },
- ],
- },
- {
- modelName: "gpt-4o-2024-11-20",
- matchPattern: "(?i)^(openai/)?(gpt-4o-2024-11-20)$",
- startDate: "2024-12-03T10:06:12.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000025,
- input_cached_tokens: 0.00000125,
- input_cache_read: 0.00000125,
- output: 0.00001,
- },
- },
- ],
- },
- {
- modelName: "gpt-4o-audio-preview",
- matchPattern: "(?i)^(openai/)?(gpt-4o-audio-preview)$",
- startDate: "2024-12-03T10:19:56.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input_text_tokens: 0.0000025,
- output_text_tokens: 0.00001,
- input_audio_tokens: 0.0001,
- input_audio: 0.0001,
- output_audio_tokens: 0.0002,
- output_audio: 0.0002,
- },
- },
- ],
- },
- {
- modelName: "gpt-4o-audio-preview-2024-10-01",
- matchPattern: "(?i)^(openai/)?(gpt-4o-audio-preview-2024-10-01)$",
- startDate: "2024-12-03T10:19:56.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input_text_tokens: 0.0000025,
- output_text_tokens: 0.00001,
- input_audio_tokens: 0.0001,
- input_audio: 0.0001,
- output_audio_tokens: 0.0002,
- output_audio: 0.0002,
- },
- },
- ],
- },
- {
- modelName: "gpt-4o-realtime-preview",
- matchPattern: "(?i)^(openai/)?(gpt-4o-realtime-preview)$",
- startDate: "2024-12-03T10:19:56.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input_text_tokens: 0.000005,
- input_cached_text_tokens: 0.0000025,
- output_text_tokens: 0.00002,
- input_audio_tokens: 0.0001,
- input_audio: 0.0001,
- input_cached_audio_tokens: 0.00002,
- output_audio_tokens: 0.0002,
- output_audio: 0.0002,
- },
- },
- ],
- },
- {
- modelName: "gpt-4o-realtime-preview-2024-10-01",
- matchPattern: "(?i)^(openai/)?(gpt-4o-realtime-preview-2024-10-01)$",
- startDate: "2024-12-03T10:19:56.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input_text_tokens: 0.000005,
- input_cached_text_tokens: 0.0000025,
- output_text_tokens: 0.00002,
- input_audio_tokens: 0.0001,
- input_audio: 0.0001,
- input_cached_audio_tokens: 0.00002,
- output_audio_tokens: 0.0002,
- output_audio: 0.0002,
- },
- },
- ],
- },
- {
- modelName: "o1",
- matchPattern: "(?i)^(openai/)?(o1)$",
- startDate: "2025-01-17T00:01:35.373Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000015,
- input_cached_tokens: 0.0000075,
- input_cache_read: 0.0000075,
- output: 0.00006,
- output_reasoning_tokens: 0.00006,
- output_reasoning: 0.00006,
- },
- },
- ],
- },
- {
- modelName: "o1-2024-12-17",
- matchPattern: "(?i)^(openai/)?(o1-2024-12-17)$",
- startDate: "2025-01-17T00:01:35.373Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000015,
- input_cached_tokens: 0.0000075,
- input_cache_read: 0.0000075,
- output: 0.00006,
- output_reasoning_tokens: 0.00006,
- output_reasoning: 0.00006,
- },
- },
- ],
- },
- {
- modelName: "o3-mini",
- matchPattern: "(?i)^(openai/)?(o3-mini)$",
- startDate: "2025-01-31T20:41:35.373Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000011,
- input_cached_tokens: 5.5e-7,
- input_cache_read: 5.5e-7,
- output: 0.0000044,
- output_reasoning_tokens: 0.0000044,
- output_reasoning: 0.0000044,
- },
- },
- ],
- },
- {
- modelName: "o3-mini-2025-01-31",
- matchPattern: "(?i)^(openai/)?(o3-mini-2025-01-31)$",
- startDate: "2025-01-31T20:41:35.373Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000011,
- input_cached_tokens: 5.5e-7,
- input_cache_read: 5.5e-7,
- output: 0.0000044,
- output_reasoning_tokens: 0.0000044,
- output_reasoning: 0.0000044,
- },
- },
- ],
- },
- {
- modelName: "gemini-2.0-flash-001",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-2.0-flash-001)(@[a-zA-Z0-9]+)?$",
- startDate: "2025-02-06T11:11:35.241Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 1e-7,
- output: 4e-7,
- },
- },
- ],
- },
- {
- modelName: "gemini-2.0-flash-lite-preview-02-05",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-2.0-flash-lite-preview-02-05)(@[a-zA-Z0-9]+)?$",
- startDate: "2025-02-06T11:11:35.241Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 7.5e-8,
- output: 3e-7,
- },
- },
- ],
- },
- {
- modelName: "claude-3.7-sonnet-20250219",
- matchPattern:
- "(?i)^(anthropic/)?(claude-3.7-sonnet-20250219|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-3.7-sonnet-20250219-v1:0|claude-3-7-sonnet-V1@20250219)$",
- startDate: "2025-02-25T09:35:39.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- input_tokens: 0.000003,
- output: 0.000015,
- output_tokens: 0.000015,
- cache_creation_input_tokens: 0.00000375,
- input_cache_creation: 0.00000375,
- input_cache_creation_5m: 0.00000375,
- input_cache_creation_1h: 0.000006,
- cache_read_input_tokens: 3e-7,
- input_cache_read: 3e-7,
- },
- },
- ],
- },
- {
- modelName: "claude-3.7-sonnet-latest",
- matchPattern: "(?i)^(anthropic/)?(claude-3-7-sonnet-latest)$",
- startDate: "2025-02-25T09:35:39.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- input_tokens: 0.000003,
- output: 0.000015,
- output_tokens: 0.000015,
- cache_creation_input_tokens: 0.00000375,
- input_cache_creation: 0.00000375,
- input_cache_creation_5m: 0.00000375,
- input_cache_creation_1h: 0.000006,
- cache_read_input_tokens: 3e-7,
- input_cache_read: 3e-7,
- },
- },
- ],
- },
- {
- modelName: "gpt-4.5-preview",
- matchPattern: "(?i)^(openai/)?(gpt-4.5-preview)$",
- startDate: "2025-02-27T21:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000075,
- input_cached_tokens: 0.0000375,
- input_cached_text_tokens: 0.0000375,
- input_cache_read: 0.0000375,
- output: 0.00015,
- },
- },
- ],
- },
- {
- modelName: "gpt-4.5-preview-2025-02-27",
- matchPattern: "(?i)^(openai/)?(gpt-4.5-preview-2025-02-27)$",
- startDate: "2025-02-27T21:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000075,
- input_cached_tokens: 0.0000375,
- input_cached_text_tokens: 0.0000375,
- input_cache_read: 0.0000375,
- output: 0.00015,
- },
- },
- ],
- },
- {
- modelName: "gpt-4.1",
- matchPattern: "(?i)^(openai/)?(gpt-4.1)$",
- startDate: "2025-04-15T10:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000002,
- input_cached_tokens: 5e-7,
- input_cached_text_tokens: 5e-7,
- input_cache_read: 5e-7,
- output: 0.000008,
- },
- },
- ],
- },
- {
- modelName: "gpt-4.1-2025-04-14",
- matchPattern: "(?i)^(openai/)?(gpt-4.1-2025-04-14)$",
- startDate: "2025-04-15T10:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000002,
- input_cached_tokens: 5e-7,
- input_cached_text_tokens: 5e-7,
- input_cache_read: 5e-7,
- output: 0.000008,
- },
- },
- ],
- },
- {
- modelName: "gpt-4.1-mini-2025-04-14",
- matchPattern: "(?i)^(openai/)?(gpt-4.1-mini-2025-04-14)$",
- startDate: "2025-04-15T10:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 4e-7,
- input_cached_tokens: 1e-7,
- input_cached_text_tokens: 1e-7,
- input_cache_read: 1e-7,
- output: 0.0000016,
- },
- },
- ],
- },
- {
- modelName: "gpt-4.1-nano-2025-04-14",
- matchPattern: "(?i)^(openai/)?(gpt-4.1-nano-2025-04-14)$",
- startDate: "2025-04-15T10:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 1e-7,
- input_cached_tokens: 2.5e-8,
- input_cached_text_tokens: 2.5e-8,
- input_cache_read: 2.5e-8,
- output: 4e-7,
- },
- },
- ],
- },
- {
- modelName: "o3",
- matchPattern: "(?i)^(openai/)?(o3)$",
- startDate: "2025-04-16T23:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000002,
- input_cached_tokens: 5e-7,
- input_cache_read: 5e-7,
- output: 0.000008,
- output_reasoning_tokens: 0.000008,
- output_reasoning: 0.000008,
- },
- },
- ],
- },
- {
- modelName: "o3-2025-04-16",
- matchPattern: "(?i)^(openai/)?(o3-2025-04-16)$",
- startDate: "2025-04-16T23:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000002,
- input_cached_tokens: 5e-7,
- input_cache_read: 5e-7,
- output: 0.000008,
- output_reasoning_tokens: 0.000008,
- output_reasoning: 0.000008,
- },
- },
- ],
- },
- {
- modelName: "o4-mini",
- matchPattern: "(?i)^(o4-mini)$",
- startDate: "2025-04-16T23:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000011,
- input_cached_tokens: 2.75e-7,
- input_cache_read: 2.75e-7,
- output: 0.0000044,
- output_reasoning_tokens: 0.0000044,
- output_reasoning: 0.0000044,
- },
- },
- ],
- },
- {
- modelName: "o4-mini-2025-04-16",
- matchPattern: "(?i)^(o4-mini-2025-04-16)$",
- startDate: "2025-04-16T23:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000011,
- input_cached_tokens: 2.75e-7,
- input_cache_read: 2.75e-7,
- output: 0.0000044,
- output_reasoning_tokens: 0.0000044,
- output_reasoning: 0.0000044,
- },
- },
- ],
- },
- {
- modelName: "gemini-2.0-flash",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-2.0-flash)(@[a-zA-Z0-9]+)?$",
- startDate: "2025-04-22T10:11:35.241Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 1e-7,
- output: 4e-7,
- },
- },
- ],
- },
- {
- modelName: "gemini-2.0-flash-lite-preview",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-2.0-flash-lite-preview)(@[a-zA-Z0-9]+)?$",
- startDate: "2025-04-22T10:11:35.241Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 7.5e-8,
- output: 3e-7,
- },
- },
- ],
- },
- {
- modelName: "gpt-4.1-nano",
- matchPattern: "(?i)^(openai/)?(gpt-4.1-nano)$",
- startDate: "2025-04-22T10:11:35.241Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 1e-7,
- input_cached_tokens: 2.5e-8,
- input_cached_text_tokens: 2.5e-8,
- input_cache_read: 2.5e-8,
- output: 4e-7,
- },
- },
- ],
- },
- {
- modelName: "gpt-4.1-mini",
- matchPattern: "(?i)^(openai/)?(gpt-4.1-mini)$",
- startDate: "2025-04-22T10:11:35.241Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 4e-7,
- input_cached_tokens: 1e-7,
- input_cached_text_tokens: 1e-7,
- input_cache_read: 1e-7,
- output: 0.0000016,
- },
- },
- ],
- },
- {
- modelName: "claude-sonnet-4-5-20250929",
- matchPattern:
- "(?i)^(anthropic/)?(claude-sonnet-4-5(-20250929)?|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-sonnet-4-5(-20250929)?-v1(:0)?|claude-sonnet-4-5-V1(@20250929)?|claude-sonnet-4-5(@20250929)?)$",
- startDate: "2025-09-29T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- input_tokens: 0.000003,
- output: 0.000015,
- output_tokens: 0.000015,
- cache_creation_input_tokens: 0.00000375,
- input_cache_creation: 0.00000375,
- input_cache_creation_5m: 0.00000375,
- input_cache_creation_1h: 0.000006,
- cache_read_input_tokens: 3e-7,
- input_cache_read: 3e-7,
- },
- },
- {
- name: "Large Context",
- isDefault: false,
- priority: 1,
- conditions: [
+ "modelName": "gpt-4o",
+ "matchPattern": "(?i)^(openai/)?(gpt-4o)$",
+ "startDate": "2024-05-13T23:15:07.670Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000025,
+ "input_cached_tokens": 0.00000125,
+ "input_cache_read": 0.00000125,
+ "output": 0.00001
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4o-2024-05-13",
+ "matchPattern": "(?i)^(openai/)?(gpt-4o-2024-05-13)$",
+ "startDate": "2024-05-13T23:15:07.670Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000005,
+ "output": 0.000015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-1106-preview",
+ "matchPattern": "(?i)^(openai/)?(gpt-4-1106-preview)$",
+ "startDate": "2024-04-23T10:37:17.092Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00001,
+ "output": 0.00003
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-turbo-vision",
+ "matchPattern": "(?i)^(openai/)?(gpt-4(-\\d{4})?-vision-preview)$",
+ "startDate": "2024-01-24T10:19:21.693Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00001,
+ "output": 0.00003
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-32k",
+ "matchPattern": "(?i)^(openai/)?(gpt-4-32k)$",
+ "startDate": "2024-01-24T10:19:21.693Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00006,
+ "output": 0.00012
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-32k-0613",
+ "matchPattern": "(?i)^(openai/)?(gpt-4-32k-0613)$",
+ "startDate": "2024-01-24T10:19:21.693Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00006,
+ "output": 0.00012
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-3.5-turbo-1106",
+ "matchPattern": "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-1106)$",
+ "startDate": "2024-01-24T10:19:21.693Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000001,
+ "output": 0.000002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-3.5-turbo-0613",
+ "matchPattern": "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-0613)$",
+ "startDate": "2024-01-24T10:19:21.693Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000015,
+ "output": 0.000002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-0613",
+ "matchPattern": "(?i)^(openai/)?(gpt-4-0613)$",
+ "startDate": "2024-01-24T10:19:21.693Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00003,
+ "output": 0.00006
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-3.5-turbo-instruct",
+ "matchPattern": "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-instruct)$",
+ "startDate": "2024-01-24T10:19:21.693Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000015,
+ "output": 0.000002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-ada-001",
+ "matchPattern": "(?i)^(text-ada-001)$",
+ "startDate": "2024-01-24T18:18:50.861Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 0.000004
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-babbage-001",
+ "matchPattern": "(?i)^(text-babbage-001)$",
+ "startDate": "2024-01-24T18:18:50.861Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-curie-001",
+ "matchPattern": "(?i)^(text-curie-001)$",
+ "startDate": "2024-01-24T18:18:50.861Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 0.00002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-davinci-001",
+ "matchPattern": "(?i)^(text-davinci-001)$",
+ "startDate": "2024-01-24T18:18:50.861Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 0.00002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-davinci-002",
+ "matchPattern": "(?i)^(text-davinci-002)$",
+ "startDate": "2024-01-24T18:18:50.861Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 0.00002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-davinci-003",
+ "matchPattern": "(?i)^(text-davinci-003)$",
+ "startDate": "2024-01-24T18:18:50.861Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 0.00002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-embedding-ada-002-v2",
+ "matchPattern": "(?i)^(text-embedding-ada-002-v2)$",
+ "startDate": "2024-01-24T18:18:50.861Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 1e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-embedding-ada-002",
+ "matchPattern": "(?i)^(text-embedding-ada-002)$",
+ "startDate": "2024-01-24T18:18:50.861Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 1e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-3.5-turbo-16k-0613",
+ "matchPattern": "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-16k-0613)$",
+ "startDate": "2024-02-03T17:29:57.350Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "output": 0.000004
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-3.5-turbo-0301",
+ "matchPattern": "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-0301)$",
+ "startDate": "2024-01-24T10:19:21.693Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000002,
+ "output": 0.000002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-32k-0314",
+ "matchPattern": "(?i)^(openai/)?(gpt-4-32k-0314)$",
+ "startDate": "2024-01-24T10:19:21.693Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00006,
+ "output": 0.00012
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-0314",
+ "matchPattern": "(?i)^(openai/)?(gpt-4-0314)$",
+ "startDate": "2024-01-24T10:19:21.693Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00003,
+ "output": 0.00006
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4",
+ "matchPattern": "(?i)^(openai/)?(gpt-4)$",
+ "startDate": "2024-01-24T10:19:21.693Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00003,
+ "output": 0.00006
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-instant-1.2",
+ "matchPattern": "(?i)^(anthropic/)?(claude-instant-1.2)$",
+ "startDate": "2024-01-30T15:44:13.447Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00000163,
+ "output": 0.00000551
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-2.0",
+ "matchPattern": "(?i)^(anthropic/)?(claude-2.0)$",
+ "startDate": "2024-01-30T15:44:13.447Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000008,
+ "output": 0.000024
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-2.1",
+ "matchPattern": "(?i)^(anthropic/)?(claude-2.1)$",
+ "startDate": "2024-01-30T15:44:13.447Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000008,
+ "output": 0.000024
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-1.3",
+ "matchPattern": "(?i)^(anthropic/)?(claude-1.3)$",
+ "startDate": "2024-01-30T15:44:13.447Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000008,
+ "output": 0.000024
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-1.2",
+ "matchPattern": "(?i)^(anthropic/)?(claude-1.2)$",
+ "startDate": "2024-01-30T15:44:13.447Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000008,
+ "output": 0.000024
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-1.1",
+ "matchPattern": "(?i)^(anthropic/)?(claude-1.1)$",
+ "startDate": "2024-01-30T15:44:13.447Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000008,
+ "output": 0.000024
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-instant-1",
+ "matchPattern": "(?i)^(anthropic/)?(claude-instant-1)$",
+ "startDate": "2024-01-30T15:44:13.447Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00000163,
+ "output": 0.00000551
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "babbage-002",
+ "matchPattern": "(?i)^(babbage-002)$",
+ "startDate": "2024-01-26T17:35:21.129Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 4e-7,
+ "output": 0.0000016
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "davinci-002",
+ "matchPattern": "(?i)^(davinci-002)$",
+ "startDate": "2024-01-26T17:35:21.129Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000006,
+ "output": 0.000012
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-embedding-3-small",
+ "matchPattern": "(?i)^(text-embedding-3-small)$",
+ "startDate": "2024-01-26T17:35:21.129Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 2e-8
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-embedding-3-large",
+ "matchPattern": "(?i)^(text-embedding-3-large)$",
+ "startDate": "2024-01-26T17:35:21.129Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 1.3e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-3.5-turbo-0125",
+ "matchPattern": "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-0125)$",
+ "startDate": "2024-01-26T17:35:21.129Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 5e-7,
+ "output": 0.0000015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-3.5-turbo",
+ "matchPattern": "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo)$",
+ "startDate": "2024-02-13T12:00:37.424Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 5e-7,
+ "output": 0.0000015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-0125-preview",
+ "matchPattern": "(?i)^(openai/)?(gpt-4-0125-preview)$",
+ "startDate": "2024-01-26T17:35:21.129Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00001,
+ "output": 0.00003
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "ft:gpt-3.5-turbo-1106",
+ "matchPattern": "(?i)^(ft:)(gpt-3.5-turbo-1106:)(.+)(:)(.*)(:)(.+)$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "output": 0.000006
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "ft:gpt-3.5-turbo-0613",
+ "matchPattern": "(?i)^(ft:)(gpt-3.5-turbo-0613:)(.+)(:)(.*)(:)(.+)$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000012,
+ "output": 0.000016
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "ft:davinci-002",
+ "matchPattern": "(?i)^(ft:)(davinci-002:)(.+)(:)(.*)(:)(.+)$$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000012,
+ "output": 0.000012
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "ft:babbage-002",
+ "matchPattern": "(?i)^(ft:)(babbage-002:)(.+)(:)(.*)(:)(.+)$$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000016,
+ "output": 0.0000016
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "chat-bison",
+ "matchPattern": "(?i)^(chat-bison)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "output": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "codechat-bison-32k",
+ "matchPattern": "(?i)^(codechat-bison-32k)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "output": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "codechat-bison",
+ "matchPattern": "(?i)^(codechat-bison)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "output": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-bison-32k",
+ "matchPattern": "(?i)^(text-bison-32k)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "output": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "chat-bison-32k",
+ "matchPattern": "(?i)^(chat-bison-32k)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "output": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-unicorn",
+ "matchPattern": "(?i)^(text-unicorn)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000025,
+ "output": 0.0000075
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "text-bison",
+ "matchPattern": "(?i)^(text-bison)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "output": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "textembedding-gecko",
+ "matchPattern": "(?i)^(textembedding-gecko)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 1e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "textembedding-gecko-multilingual",
+ "matchPattern": "(?i)^(textembedding-gecko-multilingual)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "total": 1e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "code-gecko",
+ "matchPattern": "(?i)^(code-gecko)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "output": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "code-bison",
+ "matchPattern": "(?i)^(code-bison)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "output": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "code-bison-32k",
+ "matchPattern": "(?i)^(code-bison-32k)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-01-31T13:25:02.141Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "output": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-3.5-turbo-16k",
+ "matchPattern": "(?i)^(openai/)?(gpt-)(35|3.5)(-turbo-16k)$",
+ "startDate": "2024-02-13T12:00:37.424Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 5e-7,
+ "output": 0.0000015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-turbo-preview",
+ "matchPattern": "(?i)^(openai/)?(gpt-4-turbo-preview)$",
+ "startDate": "2024-02-15T21:21:50.947Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00001,
+ "output": 0.00003
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-3-opus-20240229",
+ "matchPattern": "(?i)^(anthropic/)?(claude-3-opus-20240229|anthropic\\.claude-3-opus-20240229-v1:0|claude-3-opus@20240229)$",
+ "startDate": "2024-03-07T17:55:38.139Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000015,
+ "output": 0.000075
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-3-sonnet-20240229",
+ "matchPattern": "(?i)^(anthropic/)?(claude-3-sonnet-20240229|anthropic\\.claude-3-sonnet-20240229-v1:0|claude-3-sonnet@20240229)$",
+ "startDate": "2024-03-07T17:55:38.139Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "input_tokens": 0.000003,
+ "output": 0.000015,
+ "output_tokens": 0.000015,
+ "cache_creation_input_tokens": 0.00000375,
+ "input_cache_creation": 0.00000375,
+ "input_cache_creation_5m": 0.00000375,
+ "input_cache_creation_1h": 0.000006,
+ "cache_read_input_tokens": 3e-7,
+ "input_cache_read": 3e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-3-haiku-20240307",
+ "matchPattern": "(?i)^(anthropic/)?(claude-3-haiku-20240307|anthropic\\.claude-3-haiku-20240307-v1:0|claude-3-haiku@20240307)$",
+ "startDate": "2024-03-14T09:41:18.736Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "output": 0.00000125
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-1.0-pro-latest",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-1.0-pro-latest)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-04-11T10:27:46.517Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "output": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-1.0-pro",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-1.0-pro)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-04-11T10:27:46.517Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 1.25e-7,
+ "output": 3.75e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-1.0-pro-001",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-1.0-pro-001)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-04-11T10:27:46.517Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 1.25e-7,
+ "output": 3.75e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-pro",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-pro)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-04-11T10:27:46.517Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 1.25e-7,
+ "output": 3.75e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-1.5-pro-latest",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-1.5-pro-latest)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2024-04-11T10:27:46.517Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000025,
+ "output": 0.0000075
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-turbo-2024-04-09",
+ "matchPattern": "(?i)^(openai/)?(gpt-4-turbo-2024-04-09)$",
+ "startDate": "2024-04-23T10:37:17.092Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00001,
+ "output": 0.00003
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-turbo",
+ "matchPattern": "(?i)^(openai/)?(gpt-4-turbo)$",
+ "startDate": "2024-04-11T21:13:44.989Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00001,
+ "output": 0.00003
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4-preview",
+ "matchPattern": "(?i)^(openai/)?(gpt-4-preview)$",
+ "startDate": "2024-04-23T10:37:17.092Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00001,
+ "output": 0.00003
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-3-5-sonnet-20240620",
+ "matchPattern": "(?i)^(anthropic/)?(claude-3-5-sonnet-20240620|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-3-5-sonnet-20240620-v1:0|claude-3-5-sonnet@20240620)$",
+ "startDate": "2024-06-25T11:47:24.475Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "input_tokens": 0.000003,
+ "output": 0.000015,
+ "output_tokens": 0.000015,
+ "cache_creation_input_tokens": 0.00000375,
+ "input_cache_creation": 0.00000375,
+ "input_cache_creation_5m": 0.00000375,
+ "input_cache_creation_1h": 0.000006,
+ "cache_read_input_tokens": 3e-7,
+ "input_cache_read": 3e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4o-mini",
+ "matchPattern": "(?i)^(openai/)?(gpt-4o-mini)$",
+ "startDate": "2024-07-18T17:56:09.591Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 1.5e-7,
+ "output": 6e-7,
+ "input_cached_tokens": 7.5e-8,
+ "input_cache_read": 7.5e-8
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4o-mini-2024-07-18",
+ "matchPattern": "(?i)^(openai/)?(gpt-4o-mini-2024-07-18)$",
+ "startDate": "2024-07-18T17:56:09.591Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 1.5e-7,
+ "input_cached_tokens": 7.5e-8,
+ "input_cache_read": 7.5e-8,
+ "output": 6e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4o-2024-08-06",
+ "matchPattern": "(?i)^(openai/)?(gpt-4o-2024-08-06)$",
+ "startDate": "2024-08-07T11:54:31.298Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000025,
+ "input_cached_tokens": 0.00000125,
+ "input_cache_read": 0.00000125,
+ "output": 0.00001
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o1-preview",
+ "matchPattern": "(?i)^(openai/)?(o1-preview)$",
+ "startDate": "2024-09-13T10:01:35.373Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000015,
+ "input_cached_tokens": 0.0000075,
+ "input_cache_read": 0.0000075,
+ "output": 0.00006,
+ "output_reasoning_tokens": 0.00006,
+ "output_reasoning": 0.00006
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o1-preview-2024-09-12",
+ "matchPattern": "(?i)^(openai/)?(o1-preview-2024-09-12)$",
+ "startDate": "2024-09-13T10:01:35.373Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000015,
+ "input_cached_tokens": 0.0000075,
+ "input_cache_read": 0.0000075,
+ "output": 0.00006,
+ "output_reasoning_tokens": 0.00006,
+ "output_reasoning": 0.00006
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o1-mini",
+ "matchPattern": "(?i)^(openai/)?(o1-mini)$",
+ "startDate": "2024-09-13T10:01:35.373Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000011,
+ "input_cached_tokens": 5.5e-7,
+ "input_cache_read": 5.5e-7,
+ "output": 0.0000044,
+ "output_reasoning_tokens": 0.0000044,
+ "output_reasoning": 0.0000044
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o1-mini-2024-09-12",
+ "matchPattern": "(?i)^(openai/)?(o1-mini-2024-09-12)$",
+ "startDate": "2024-09-13T10:01:35.373Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000011,
+ "input_cached_tokens": 5.5e-7,
+ "input_cache_read": 5.5e-7,
+ "output": 0.0000044,
+ "output_reasoning_tokens": 0.0000044,
+ "output_reasoning": 0.0000044
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-3.5-sonnet-20241022",
+ "matchPattern": "(?i)^(anthropic/)?(claude-3-5-sonnet-20241022|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-3-5-sonnet-20241022-v2:0|claude-3-5-sonnet-V2@20241022)$",
+ "startDate": "2024-10-22T18:48:01.676Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "input_tokens": 0.000003,
+ "output": 0.000015,
+ "output_tokens": 0.000015,
+ "cache_creation_input_tokens": 0.00000375,
+ "input_cache_creation": 0.00000375,
+ "input_cache_creation_5m": 0.00000375,
+ "input_cache_creation_1h": 0.000006,
+ "cache_read_input_tokens": 3e-7,
+ "input_cache_read": 3e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-3.5-sonnet-latest",
+ "matchPattern": "(?i)^(anthropic/)?(claude-3-5-sonnet-latest)$",
+ "startDate": "2024-10-22T18:48:01.676Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "input_tokens": 0.000003,
+ "output": 0.000015,
+ "output_tokens": 0.000015,
+ "cache_creation_input_tokens": 0.00000375,
+ "input_cache_creation": 0.00000375,
+ "input_cache_creation_5m": 0.00000375,
+ "input_cache_creation_1h": 0.000006,
+ "cache_read_input_tokens": 3e-7,
+ "input_cache_read": 3e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-3-5-haiku-20241022",
+ "matchPattern": "(?i)^(anthropic/)?(claude-3-5-haiku-20241022|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-3-5-haiku-20241022-v1:0|claude-3-5-haiku-V1@20241022)$",
+ "startDate": "2024-11-05T10:30:50.566Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 8e-7,
+ "input_tokens": 8e-7,
+ "output": 0.000004,
+ "output_tokens": 0.000004,
+ "cache_creation_input_tokens": 0.000001,
+ "input_cache_creation": 0.000001,
+ "input_cache_creation_5m": 0.000001,
+ "input_cache_creation_1h": 0.0000016,
+ "cache_read_input_tokens": 8e-8,
+ "input_cache_read": 8e-8
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-3.5-haiku-latest",
+ "matchPattern": "(?i)^(anthropic/)?(claude-3-5-haiku-latest)$",
+ "startDate": "2024-11-05T10:30:50.566Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 8e-7,
+ "input_tokens": 8e-7,
+ "output": 0.000004,
+ "output_tokens": 0.000004,
+ "cache_creation_input_tokens": 0.000001,
+ "input_cache_creation": 0.000001,
+ "input_cache_creation_5m": 0.000001,
+ "input_cache_creation_1h": 0.0000016,
+ "cache_read_input_tokens": 8e-8,
+ "input_cache_read": 8e-8
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "chatgpt-4o-latest",
+ "matchPattern": "(?i)^(chatgpt-4o-latest)$",
+ "startDate": "2024-11-25T12:47:17.504Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000005,
+ "output": 0.000015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4o-2024-11-20",
+ "matchPattern": "(?i)^(openai/)?(gpt-4o-2024-11-20)$",
+ "startDate": "2024-12-03T10:06:12.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000025,
+ "input_cached_tokens": 0.00000125,
+ "input_cache_read": 0.00000125,
+ "output": 0.00001
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4o-audio-preview",
+ "matchPattern": "(?i)^(openai/)?(gpt-4o-audio-preview)$",
+ "startDate": "2024-12-03T10:19:56.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input_text_tokens": 0.0000025,
+ "output_text_tokens": 0.00001,
+ "input_audio_tokens": 0.0001,
+ "input_audio": 0.0001,
+ "output_audio_tokens": 0.0002,
+ "output_audio": 0.0002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4o-audio-preview-2024-10-01",
+ "matchPattern": "(?i)^(openai/)?(gpt-4o-audio-preview-2024-10-01)$",
+ "startDate": "2024-12-03T10:19:56.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input_text_tokens": 0.0000025,
+ "output_text_tokens": 0.00001,
+ "input_audio_tokens": 0.0001,
+ "input_audio": 0.0001,
+ "output_audio_tokens": 0.0002,
+ "output_audio": 0.0002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4o-realtime-preview",
+ "matchPattern": "(?i)^(openai/)?(gpt-4o-realtime-preview)$",
+ "startDate": "2024-12-03T10:19:56.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input_text_tokens": 0.000005,
+ "input_cached_text_tokens": 0.0000025,
+ "output_text_tokens": 0.00002,
+ "input_audio_tokens": 0.0001,
+ "input_audio": 0.0001,
+ "input_cached_audio_tokens": 0.00002,
+ "output_audio_tokens": 0.0002,
+ "output_audio": 0.0002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4o-realtime-preview-2024-10-01",
+ "matchPattern": "(?i)^(openai/)?(gpt-4o-realtime-preview-2024-10-01)$",
+ "startDate": "2024-12-03T10:19:56.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input_text_tokens": 0.000005,
+ "input_cached_text_tokens": 0.0000025,
+ "output_text_tokens": 0.00002,
+ "input_audio_tokens": 0.0001,
+ "input_audio": 0.0001,
+ "input_cached_audio_tokens": 0.00002,
+ "output_audio_tokens": 0.0002,
+ "output_audio": 0.0002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o1",
+ "matchPattern": "(?i)^(openai/)?(o1)$",
+ "startDate": "2025-01-17T00:01:35.373Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000015,
+ "input_cached_tokens": 0.0000075,
+ "input_cache_read": 0.0000075,
+ "output": 0.00006,
+ "output_reasoning_tokens": 0.00006,
+ "output_reasoning": 0.00006
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o1-2024-12-17",
+ "matchPattern": "(?i)^(openai/)?(o1-2024-12-17)$",
+ "startDate": "2025-01-17T00:01:35.373Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000015,
+ "input_cached_tokens": 0.0000075,
+ "input_cache_read": 0.0000075,
+ "output": 0.00006,
+ "output_reasoning_tokens": 0.00006,
+ "output_reasoning": 0.00006
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o3-mini",
+ "matchPattern": "(?i)^(openai/)?(o3-mini)$",
+ "startDate": "2025-01-31T20:41:35.373Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000011,
+ "input_cached_tokens": 5.5e-7,
+ "input_cache_read": 5.5e-7,
+ "output": 0.0000044,
+ "output_reasoning_tokens": 0.0000044,
+ "output_reasoning": 0.0000044
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o3-mini-2025-01-31",
+ "matchPattern": "(?i)^(openai/)?(o3-mini-2025-01-31)$",
+ "startDate": "2025-01-31T20:41:35.373Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000011,
+ "input_cached_tokens": 5.5e-7,
+ "input_cache_read": 5.5e-7,
+ "output": 0.0000044,
+ "output_reasoning_tokens": 0.0000044,
+ "output_reasoning": 0.0000044
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-2.0-flash-001",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-2.0-flash-001)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2025-02-06T11:11:35.241Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 1e-7,
+ "output": 4e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-2.0-flash-lite-preview-02-05",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-2.0-flash-lite-preview-02-05)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2025-02-06T11:11:35.241Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 7.5e-8,
+ "output": 3e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-3.7-sonnet-20250219",
+ "matchPattern": "(?i)^(anthropic/)?(claude-3.7-sonnet-20250219|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-3.7-sonnet-20250219-v1:0|claude-3-7-sonnet-V1@20250219)$",
+ "startDate": "2025-02-25T09:35:39.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "input_tokens": 0.000003,
+ "output": 0.000015,
+ "output_tokens": 0.000015,
+ "cache_creation_input_tokens": 0.00000375,
+ "input_cache_creation": 0.00000375,
+ "input_cache_creation_5m": 0.00000375,
+ "input_cache_creation_1h": 0.000006,
+ "cache_read_input_tokens": 3e-7,
+ "input_cache_read": 3e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-3.7-sonnet-latest",
+ "matchPattern": "(?i)^(anthropic/)?(claude-3-7-sonnet-latest)$",
+ "startDate": "2025-02-25T09:35:39.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "input_tokens": 0.000003,
+ "output": 0.000015,
+ "output_tokens": 0.000015,
+ "cache_creation_input_tokens": 0.00000375,
+ "input_cache_creation": 0.00000375,
+ "input_cache_creation_5m": 0.00000375,
+ "input_cache_creation_1h": 0.000006,
+ "cache_read_input_tokens": 3e-7,
+ "input_cache_read": 3e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4.5-preview",
+ "matchPattern": "(?i)^(openai/)?(gpt-4.5-preview)$",
+ "startDate": "2025-02-27T21:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000075,
+ "input_cached_tokens": 0.0000375,
+ "input_cached_text_tokens": 0.0000375,
+ "input_cache_read": 0.0000375,
+ "output": 0.00015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4.5-preview-2025-02-27",
+ "matchPattern": "(?i)^(openai/)?(gpt-4.5-preview-2025-02-27)$",
+ "startDate": "2025-02-27T21:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000075,
+ "input_cached_tokens": 0.0000375,
+ "input_cached_text_tokens": 0.0000375,
+ "input_cache_read": 0.0000375,
+ "output": 0.00015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4.1",
+ "matchPattern": "(?i)^(openai/)?(gpt-4.1)$",
+ "startDate": "2025-04-15T10:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000002,
+ "input_cached_tokens": 5e-7,
+ "input_cached_text_tokens": 5e-7,
+ "input_cache_read": 5e-7,
+ "output": 0.000008
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4.1-2025-04-14",
+ "matchPattern": "(?i)^(openai/)?(gpt-4.1-2025-04-14)$",
+ "startDate": "2025-04-15T10:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000002,
+ "input_cached_tokens": 5e-7,
+ "input_cached_text_tokens": 5e-7,
+ "input_cache_read": 5e-7,
+ "output": 0.000008
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4.1-mini-2025-04-14",
+ "matchPattern": "(?i)^(openai/)?(gpt-4.1-mini-2025-04-14)$",
+ "startDate": "2025-04-15T10:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 4e-7,
+ "input_cached_tokens": 1e-7,
+ "input_cached_text_tokens": 1e-7,
+ "input_cache_read": 1e-7,
+ "output": 0.0000016
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4.1-nano-2025-04-14",
+ "matchPattern": "(?i)^(openai/)?(gpt-4.1-nano-2025-04-14)$",
+ "startDate": "2025-04-15T10:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 1e-7,
+ "input_cached_tokens": 2.5e-8,
+ "input_cached_text_tokens": 2.5e-8,
+ "input_cache_read": 2.5e-8,
+ "output": 4e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o3",
+ "matchPattern": "(?i)^(openai/)?(o3)$",
+ "startDate": "2025-04-16T23:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000002,
+ "input_cached_tokens": 5e-7,
+ "input_cache_read": 5e-7,
+ "output": 0.000008,
+ "output_reasoning_tokens": 0.000008,
+ "output_reasoning": 0.000008
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o3-2025-04-16",
+ "matchPattern": "(?i)^(openai/)?(o3-2025-04-16)$",
+ "startDate": "2025-04-16T23:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000002,
+ "input_cached_tokens": 5e-7,
+ "input_cache_read": 5e-7,
+ "output": 0.000008,
+ "output_reasoning_tokens": 0.000008,
+ "output_reasoning": 0.000008
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o4-mini",
+ "matchPattern": "(?i)^(o4-mini)$",
+ "startDate": "2025-04-16T23:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000011,
+ "input_cached_tokens": 2.75e-7,
+ "input_cache_read": 2.75e-7,
+ "output": 0.0000044,
+ "output_reasoning_tokens": 0.0000044,
+ "output_reasoning": 0.0000044
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o4-mini-2025-04-16",
+ "matchPattern": "(?i)^(o4-mini-2025-04-16)$",
+ "startDate": "2025-04-16T23:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000011,
+ "input_cached_tokens": 2.75e-7,
+ "input_cache_read": 2.75e-7,
+ "output": 0.0000044,
+ "output_reasoning_tokens": 0.0000044,
+ "output_reasoning": 0.0000044
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-2.0-flash",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-2.0-flash)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2025-04-22T10:11:35.241Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 1e-7,
+ "output": 4e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-2.0-flash-lite-preview",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-2.0-flash-lite-preview)(@[a-zA-Z0-9]+)?$",
+ "startDate": "2025-04-22T10:11:35.241Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 7.5e-8,
+ "output": 3e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4.1-nano",
+ "matchPattern": "(?i)^(openai/)?(gpt-4.1-nano)$",
+ "startDate": "2025-04-22T10:11:35.241Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 1e-7,
+ "input_cached_tokens": 2.5e-8,
+ "input_cached_text_tokens": 2.5e-8,
+ "input_cache_read": 2.5e-8,
+ "output": 4e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-4.1-mini",
+ "matchPattern": "(?i)^(openai/)?(gpt-4.1-mini)$",
+ "startDate": "2025-04-22T10:11:35.241Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 4e-7,
+ "input_cached_tokens": 1e-7,
+ "input_cached_text_tokens": 1e-7,
+ "input_cache_read": 1e-7,
+ "output": 0.0000016
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-sonnet-4-5-20250929",
+ "matchPattern": "(?i)^(anthropic/)?(claude-sonnet-4-5(-20250929)?|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-sonnet-4-5(-20250929)?-v1(:0)?|claude-sonnet-4-5-V1(@20250929)?|claude-sonnet-4-5(@20250929)?)$",
+ "startDate": "2025-09-29T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "input_tokens": 0.000003,
+ "output": 0.000015,
+ "output_tokens": 0.000015,
+ "cache_creation_input_tokens": 0.00000375,
+ "input_cache_creation": 0.00000375,
+ "input_cache_creation_5m": 0.00000375,
+ "input_cache_creation_1h": 0.000006,
+ "cache_read_input_tokens": 3e-7,
+ "input_cache_read": 3e-7
+ }
+ },
+ {
+ "name": "Large Context",
+ "isDefault": false,
+ "priority": 1,
+ "conditions": [
{
- usageDetailPattern: "input",
- operator: "gt",
- value: 200000,
- },
+ "usageDetailPattern": "input",
+ "operator": "gt",
+ "value": 200000
+ }
],
- prices: {
- input: 0.000006,
- input_tokens: 0.000006,
- output: 0.0000225,
- output_tokens: 0.0000225,
- cache_creation_input_tokens: 0.0000075,
- input_cache_creation: 0.0000075,
- input_cache_creation_5m: 0.0000075,
- input_cache_creation_1h: 0.000012,
- cache_read_input_tokens: 6e-7,
- input_cache_read: 6e-7,
- },
- },
- ],
- },
- {
- modelName: "claude-sonnet-4-20250514",
- matchPattern:
- "(?i)^(anthropic/)?(claude-sonnet-4(-20250514)?|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-sonnet-4(-20250514)?-v1(:0)?|claude-sonnet-4-V1(@20250514)?|claude-sonnet-4(@20250514)?)$",
- startDate: "2025-05-22T17:09:02.131Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- input_tokens: 0.000003,
- output: 0.000015,
- output_tokens: 0.000015,
- cache_creation_input_tokens: 0.00000375,
- input_cache_creation: 0.00000375,
- input_cache_creation_5m: 0.00000375,
- input_cache_creation_1h: 0.000006,
- cache_read_input_tokens: 3e-7,
- input_cache_read: 3e-7,
- },
- },
- ],
- },
- {
- modelName: "claude-sonnet-4-latest",
- matchPattern: "(?i)^(anthropic/)?(claude-sonnet-4-latest)$",
- startDate: "2025-05-22T17:09:02.131Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- input_tokens: 0.000003,
- output: 0.000015,
- output_tokens: 0.000015,
- cache_creation_input_tokens: 0.00000375,
- input_cache_creation: 0.00000375,
- input_cache_creation_5m: 0.00000375,
- input_cache_creation_1h: 0.000006,
- cache_read_input_tokens: 3e-7,
- input_cache_read: 3e-7,
- },
- },
- ],
- },
- {
- modelName: "claude-opus-4-20250514",
- matchPattern:
- "(?i)^(anthropic/)?(claude-opus-4(-20250514)?|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-opus-4(-20250514)?-v1(:0)?|claude-opus-4(@20250514)?)$",
- startDate: "2025-05-22T17:09:02.131Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000015,
- input_tokens: 0.000015,
- output: 0.000075,
- output_tokens: 0.000075,
- cache_creation_input_tokens: 0.00001875,
- input_cache_creation: 0.00001875,
- input_cache_creation_5m: 0.00001875,
- input_cache_creation_1h: 0.00003,
- cache_read_input_tokens: 0.0000015,
- input_cache_read: 0.0000015,
- },
- },
- ],
- },
- {
- modelName: "o3-pro",
- matchPattern: "(?i)^(openai/)?(o3-pro)$",
- startDate: "2025-06-10T22:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00002,
- output: 0.00008,
- output_reasoning_tokens: 0.00008,
- output_reasoning: 0.00008,
- },
- },
- ],
- },
- {
- modelName: "o3-pro-2025-06-10",
- matchPattern: "(?i)^(openai/)?(o3-pro-2025-06-10)$",
- startDate: "2025-06-10T22:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00002,
- output: 0.00008,
- output_reasoning_tokens: 0.00008,
- output_reasoning: 0.00008,
- },
- },
- ],
- },
- {
- modelName: "o1-pro",
- matchPattern: "(?i)^(openai/)?(o1-pro)$",
- startDate: "2025-06-10T22:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00015,
- output: 0.0006,
- output_reasoning_tokens: 0.0006,
- output_reasoning: 0.0006,
- },
- },
- ],
- },
- {
- modelName: "o1-pro-2025-03-19",
- matchPattern: "(?i)^(openai/)?(o1-pro-2025-03-19)$",
- startDate: "2025-06-10T22:26:54.132Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00015,
- output: 0.0006,
- output_reasoning_tokens: 0.0006,
- output_reasoning: 0.0006,
- },
- },
- ],
- },
- {
- modelName: "gemini-2.5-flash",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-2.5-flash)$",
- startDate: "2025-07-03T13:44:06.964Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 3e-7,
- input_text: 3e-7,
- input_modality_1: 3e-7,
- prompt_token_count: 3e-7,
- promptTokenCount: 3e-7,
- input_cached_tokens: 3e-8,
- cached_content_token_count: 3e-8,
- output: 0.0000025,
- output_text: 0.0000025,
- output_modality_1: 0.0000025,
- candidates_token_count: 0.0000025,
- candidatesTokenCount: 0.0000025,
- thoughtsTokenCount: 0.0000025,
- thoughts_token_count: 0.0000025,
- output_reasoning: 0.0000025,
- input_audio_tokens: 0.000001,
- },
- },
- ],
- },
- {
- modelName: "gemini-2.5-flash-lite",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-2.5-flash-lite)$",
- startDate: "2025-07-03T13:44:06.964Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 1e-7,
- input_text: 1e-7,
- input_modality_1: 1e-7,
- prompt_token_count: 1e-7,
- promptTokenCount: 1e-7,
- input_cached_tokens: 2.5e-8,
- cached_content_token_count: 2.5e-8,
- output: 4e-7,
- output_text: 4e-7,
- output_modality_1: 4e-7,
- candidates_token_count: 4e-7,
- candidatesTokenCount: 4e-7,
- thoughtsTokenCount: 4e-7,
- thoughts_token_count: 4e-7,
- output_reasoning: 4e-7,
- input_audio_tokens: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "claude-opus-4-1-20250805",
- matchPattern:
- "(?i)^(anthropic/)?(claude-opus-4-1(-20250805)?|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-opus-4-1(-20250805)?-v1(:0)?|claude-opus-4-1(@20250805)?)$",
- startDate: "2025-08-05T15:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000015,
- input_tokens: 0.000015,
- output: 0.000075,
- output_tokens: 0.000075,
- cache_creation_input_tokens: 0.00001875,
- input_cache_creation: 0.00001875,
- input_cache_creation_5m: 0.00001875,
- input_cache_creation_1h: 0.00003,
- cache_read_input_tokens: 0.0000015,
- input_cache_read: 0.0000015,
- },
- },
- ],
- },
- {
- modelName: "gpt-5",
- matchPattern: "(?i)^(openai/)?(gpt-5)$",
- startDate: "2025-08-07T16:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00000125,
- input_cached_tokens: 1.25e-7,
- output: 0.00001,
- input_cache_read: 1.25e-7,
- output_reasoning_tokens: 0.00001,
- output_reasoning: 0.00001,
- },
- },
- ],
- },
- {
- modelName: "gpt-5-2025-08-07",
- matchPattern: "(?i)^(openai/)?(gpt-5-2025-08-07)$",
- startDate: "2025-08-11T08:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00000125,
- input_cached_tokens: 1.25e-7,
- output: 0.00001,
- input_cache_read: 1.25e-7,
- output_reasoning_tokens: 0.00001,
- output_reasoning: 0.00001,
- },
- },
- ],
- },
- {
- modelName: "gpt-5-mini",
- matchPattern: "(?i)^(openai/)?(gpt-5-mini)$",
- startDate: "2025-08-07T16:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- input_cached_tokens: 2.5e-8,
- output: 0.000002,
- input_cache_read: 2.5e-8,
- output_reasoning_tokens: 0.000002,
- output_reasoning: 0.000002,
- },
- },
- ],
- },
- {
- modelName: "gpt-5-mini-2025-08-07",
- matchPattern: "(?i)^(openai/)?(gpt-5-mini-2025-08-07)$",
- startDate: "2025-08-11T08:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- input_cached_tokens: 2.5e-8,
- output: 0.000002,
- input_cache_read: 2.5e-8,
- output_reasoning_tokens: 0.000002,
- output_reasoning: 0.000002,
- },
- },
- ],
- },
- {
- modelName: "gpt-5-nano",
- matchPattern: "(?i)^(openai/)?(gpt-5-nano)$",
- startDate: "2025-08-07T16:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 5e-8,
- input_cached_tokens: 5e-9,
- output: 4e-7,
- input_cache_read: 5e-9,
- output_reasoning_tokens: 4e-7,
- output_reasoning: 4e-7,
- },
- },
- ],
- },
- {
- modelName: "gpt-5-nano-2025-08-07",
- matchPattern: "(?i)^(openai/)?(gpt-5-nano-2025-08-07)$",
- startDate: "2025-08-11T08:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 5e-8,
- input_cached_tokens: 5e-9,
- output: 4e-7,
- input_cache_read: 5e-9,
- output_reasoning_tokens: 4e-7,
- output_reasoning: 4e-7,
- },
- },
- ],
- },
- {
- modelName: "gpt-5-chat-latest",
- matchPattern: "(?i)^(openai/)?(gpt-5-chat-latest)$",
- startDate: "2025-08-07T16:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00000125,
- input_cached_tokens: 1.25e-7,
- output: 0.00001,
- input_cache_read: 1.25e-7,
- output_reasoning_tokens: 0.00001,
- output_reasoning: 0.00001,
- },
- },
- ],
- },
- {
- modelName: "gpt-5-pro",
- matchPattern: "(?i)^(openai/)?(gpt-5-pro)$",
- startDate: "2025-10-07T08:03:54.727Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000015,
- output: 0.00012,
- output_reasoning_tokens: 0.00012,
- output_reasoning: 0.00012,
- },
- },
- ],
- },
- {
- modelName: "gpt-5-pro-2025-10-06",
- matchPattern: "(?i)^(openai/)?(gpt-5-pro-2025-10-06)$",
- startDate: "2025-10-07T08:03:54.727Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000015,
- output: 0.00012,
- output_reasoning_tokens: 0.00012,
- output_reasoning: 0.00012,
- },
- },
- ],
- },
- {
- modelName: "claude-haiku-4-5-20251001",
- matchPattern:
- "(?i)^(anthropic/)?(claude-haiku-4-5-20251001|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-haiku-4-5-20251001-v1:0|claude-4-5-haiku@20251001)$",
- startDate: "2025-10-16T08:20:44.558Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000001,
- input_tokens: 0.000001,
- output: 0.000005,
- output_tokens: 0.000005,
- cache_creation_input_tokens: 0.00000125,
- input_cache_creation: 0.00000125,
- input_cache_creation_5m: 0.00000125,
- input_cache_creation_1h: 0.000002,
- cache_read_input_tokens: 1e-7,
- input_cache_read: 1e-7,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.1",
- matchPattern: "(?i)^(openai/)?(gpt-5.1)$",
- startDate: "2025-11-14T08:57:23.481Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00000125,
- input_cached_tokens: 1.25e-7,
- output: 0.00001,
- input_cache_read: 1.25e-7,
- output_reasoning_tokens: 0.00001,
- output_reasoning: 0.00001,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.1-2025-11-13",
- matchPattern: "(?i)^(openai/)?(gpt-5.1-2025-11-13)$",
- startDate: "2025-11-14T08:57:23.481Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00000125,
- input_cached_tokens: 1.25e-7,
- output: 0.00001,
- input_cache_read: 1.25e-7,
- output_reasoning_tokens: 0.00001,
- output_reasoning: 0.00001,
- },
- },
- ],
- },
- {
- modelName: "claude-opus-4-5-20251101",
- matchPattern:
- "(?i)^(anthropic/)?(claude-opus-4-5(-20251101)?|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-opus-4-5(-20251101)?-v1(:0)?|claude-opus-4-5(@20251101)?)$",
- startDate: "2025-11-24T20:53:27.571Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000005,
- input_tokens: 0.000005,
- output: 0.000025,
- output_tokens: 0.000025,
- cache_creation_input_tokens: 0.00000625,
- input_cache_creation: 0.00000625,
- input_cache_creation_5m: 0.00000625,
- input_cache_creation_1h: 0.00001,
- cache_read_input_tokens: 5e-7,
- input_cache_read: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "claude-sonnet-4-6",
- matchPattern:
- "(?i)^(anthropic\\/)?(claude-sonnet-4-6|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-sonnet-4-6(-v1(:0)?)?|claude-sonnet-4-6)$",
- startDate: "2026-02-18T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000003,
- input_tokens: 0.000003,
- output: 0.000015,
- output_tokens: 0.000015,
- cache_creation_input_tokens: 0.00000375,
- input_cache_creation: 0.00000375,
- input_cache_creation_5m: 0.00000375,
- input_cache_creation_1h: 0.000006,
- cache_read_input_tokens: 3e-7,
- input_cache_read: 3e-7,
- },
- },
- {
- name: "Large Context",
- isDefault: false,
- priority: 1,
- conditions: [
+ "prices": {
+ "input": 0.000006,
+ "input_tokens": 0.000006,
+ "output": 0.0000225,
+ "output_tokens": 0.0000225,
+ "cache_creation_input_tokens": 0.0000075,
+ "input_cache_creation": 0.0000075,
+ "input_cache_creation_5m": 0.0000075,
+ "input_cache_creation_1h": 0.000012,
+ "cache_read_input_tokens": 6e-7,
+ "input_cache_read": 6e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-sonnet-4-20250514",
+ "matchPattern": "(?i)^(anthropic/)?(claude-sonnet-4(-20250514)?|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-sonnet-4(-20250514)?-v1(:0)?|claude-sonnet-4-V1(@20250514)?|claude-sonnet-4(@20250514)?)$",
+ "startDate": "2025-05-22T17:09:02.131Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "input_tokens": 0.000003,
+ "output": 0.000015,
+ "output_tokens": 0.000015,
+ "cache_creation_input_tokens": 0.00000375,
+ "input_cache_creation": 0.00000375,
+ "input_cache_creation_5m": 0.00000375,
+ "input_cache_creation_1h": 0.000006,
+ "cache_read_input_tokens": 3e-7,
+ "input_cache_read": 3e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-sonnet-4-latest",
+ "matchPattern": "(?i)^(anthropic/)?(claude-sonnet-4-latest)$",
+ "startDate": "2025-05-22T17:09:02.131Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "input_tokens": 0.000003,
+ "output": 0.000015,
+ "output_tokens": 0.000015,
+ "cache_creation_input_tokens": 0.00000375,
+ "input_cache_creation": 0.00000375,
+ "input_cache_creation_5m": 0.00000375,
+ "input_cache_creation_1h": 0.000006,
+ "cache_read_input_tokens": 3e-7,
+ "input_cache_read": 3e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-opus-4-20250514",
+ "matchPattern": "(?i)^(anthropic/)?(claude-opus-4(-20250514)?|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-opus-4(-20250514)?-v1(:0)?|claude-opus-4(@20250514)?)$",
+ "startDate": "2025-05-22T17:09:02.131Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000015,
+ "input_tokens": 0.000015,
+ "output": 0.000075,
+ "output_tokens": 0.000075,
+ "cache_creation_input_tokens": 0.00001875,
+ "input_cache_creation": 0.00001875,
+ "input_cache_creation_5m": 0.00001875,
+ "input_cache_creation_1h": 0.00003,
+ "cache_read_input_tokens": 0.0000015,
+ "input_cache_read": 0.0000015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o3-pro",
+ "matchPattern": "(?i)^(openai/)?(o3-pro)$",
+ "startDate": "2025-06-10T22:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00002,
+ "output": 0.00008,
+ "output_reasoning_tokens": 0.00008,
+ "output_reasoning": 0.00008
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o3-pro-2025-06-10",
+ "matchPattern": "(?i)^(openai/)?(o3-pro-2025-06-10)$",
+ "startDate": "2025-06-10T22:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00002,
+ "output": 0.00008,
+ "output_reasoning_tokens": 0.00008,
+ "output_reasoning": 0.00008
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o1-pro",
+ "matchPattern": "(?i)^(openai/)?(o1-pro)$",
+ "startDate": "2025-06-10T22:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00015,
+ "output": 0.0006,
+ "output_reasoning_tokens": 0.0006,
+ "output_reasoning": 0.0006
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "o1-pro-2025-03-19",
+ "matchPattern": "(?i)^(openai/)?(o1-pro-2025-03-19)$",
+ "startDate": "2025-06-10T22:26:54.132Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00015,
+ "output": 0.0006,
+ "output_reasoning_tokens": 0.0006,
+ "output_reasoning": 0.0006
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-2.5-flash",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-2.5-flash)$",
+ "startDate": "2025-07-03T13:44:06.964Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 3e-7,
+ "input_text": 3e-7,
+ "input_modality_1": 3e-7,
+ "prompt_token_count": 3e-7,
+ "promptTokenCount": 3e-7,
+ "input_cached_tokens": 3e-8,
+ "cached_content_token_count": 3e-8,
+ "output": 0.0000025,
+ "output_text": 0.0000025,
+ "output_modality_1": 0.0000025,
+ "candidates_token_count": 0.0000025,
+ "candidatesTokenCount": 0.0000025,
+ "thoughtsTokenCount": 0.0000025,
+ "thoughts_token_count": 0.0000025,
+ "output_reasoning": 0.0000025,
+ "input_audio_tokens": 0.000001
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-2.5-flash-lite",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-2.5-flash-lite)$",
+ "startDate": "2025-07-03T13:44:06.964Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 1e-7,
+ "input_text": 1e-7,
+ "input_modality_1": 1e-7,
+ "prompt_token_count": 1e-7,
+ "promptTokenCount": 1e-7,
+ "input_cached_tokens": 2.5e-8,
+ "cached_content_token_count": 2.5e-8,
+ "output": 4e-7,
+ "output_text": 4e-7,
+ "output_modality_1": 4e-7,
+ "candidates_token_count": 4e-7,
+ "candidatesTokenCount": 4e-7,
+ "thoughtsTokenCount": 4e-7,
+ "thoughts_token_count": 4e-7,
+ "output_reasoning": 4e-7,
+ "input_audio_tokens": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-opus-4-1-20250805",
+ "matchPattern": "(?i)^(anthropic/)?(claude-opus-4-1(-20250805)?|(eu\\.|us\\.|apac\\.)?anthropic\\.claude-opus-4-1(-20250805)?-v1(:0)?|claude-opus-4-1(@20250805)?)$",
+ "startDate": "2025-08-05T15:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000015,
+ "input_tokens": 0.000015,
+ "output": 0.000075,
+ "output_tokens": 0.000075,
+ "cache_creation_input_tokens": 0.00001875,
+ "input_cache_creation": 0.00001875,
+ "input_cache_creation_5m": 0.00001875,
+ "input_cache_creation_1h": 0.00003,
+ "cache_read_input_tokens": 0.0000015,
+ "input_cache_read": 0.0000015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5",
+ "matchPattern": "(?i)^(openai/)?(gpt-5)$",
+ "startDate": "2025-08-07T16:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00000125,
+ "input_cached_tokens": 1.25e-7,
+ "output": 0.00001,
+ "input_cache_read": 1.25e-7,
+ "output_reasoning_tokens": 0.00001,
+ "output_reasoning": 0.00001
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5-2025-08-07",
+ "matchPattern": "(?i)^(openai/)?(gpt-5-2025-08-07)$",
+ "startDate": "2025-08-11T08:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00000125,
+ "input_cached_tokens": 1.25e-7,
+ "output": 0.00001,
+ "input_cache_read": 1.25e-7,
+ "output_reasoning_tokens": 0.00001,
+ "output_reasoning": 0.00001
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5-mini",
+ "matchPattern": "(?i)^(openai/)?(gpt-5-mini)$",
+ "startDate": "2025-08-07T16:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "input_cached_tokens": 2.5e-8,
+ "output": 0.000002,
+ "input_cache_read": 2.5e-8,
+ "output_reasoning_tokens": 0.000002,
+ "output_reasoning": 0.000002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5-mini-2025-08-07",
+ "matchPattern": "(?i)^(openai/)?(gpt-5-mini-2025-08-07)$",
+ "startDate": "2025-08-11T08:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "input_cached_tokens": 2.5e-8,
+ "output": 0.000002,
+ "input_cache_read": 2.5e-8,
+ "output_reasoning_tokens": 0.000002,
+ "output_reasoning": 0.000002
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5-nano",
+ "matchPattern": "(?i)^(openai/)?(gpt-5-nano)$",
+ "startDate": "2025-08-07T16:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 5e-8,
+ "input_cached_tokens": 5e-9,
+ "output": 4e-7,
+ "input_cache_read": 5e-9,
+ "output_reasoning_tokens": 4e-7,
+ "output_reasoning": 4e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5-nano-2025-08-07",
+ "matchPattern": "(?i)^(openai/)?(gpt-5-nano-2025-08-07)$",
+ "startDate": "2025-08-11T08:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 5e-8,
+ "input_cached_tokens": 5e-9,
+ "output": 4e-7,
+ "input_cache_read": 5e-9,
+ "output_reasoning_tokens": 4e-7,
+ "output_reasoning": 4e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5-chat-latest",
+ "matchPattern": "(?i)^(openai/)?(gpt-5-chat-latest)$",
+ "startDate": "2025-08-07T16:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00000125,
+ "input_cached_tokens": 1.25e-7,
+ "output": 0.00001,
+ "input_cache_read": 1.25e-7,
+ "output_reasoning_tokens": 0.00001,
+ "output_reasoning": 0.00001
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5-pro",
+ "matchPattern": "(?i)^(openai/)?(gpt-5-pro)$",
+ "startDate": "2025-10-07T08:03:54.727Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000015,
+ "output": 0.00012,
+ "output_reasoning_tokens": 0.00012,
+ "output_reasoning": 0.00012
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5-pro-2025-10-06",
+ "matchPattern": "(?i)^(openai/)?(gpt-5-pro-2025-10-06)$",
+ "startDate": "2025-10-07T08:03:54.727Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000015,
+ "output": 0.00012,
+ "output_reasoning_tokens": 0.00012,
+ "output_reasoning": 0.00012
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-haiku-4-5-20251001",
+ "matchPattern": "(?i)^(anthropic/)?(claude-haiku-4-5-20251001|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-haiku-4-5-20251001-v1:0|claude-4-5-haiku@20251001)$",
+ "startDate": "2025-10-16T08:20:44.558Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000001,
+ "input_tokens": 0.000001,
+ "output": 0.000005,
+ "output_tokens": 0.000005,
+ "cache_creation_input_tokens": 0.00000125,
+ "input_cache_creation": 0.00000125,
+ "input_cache_creation_5m": 0.00000125,
+ "input_cache_creation_1h": 0.000002,
+ "cache_read_input_tokens": 1e-7,
+ "input_cache_read": 1e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.1",
+ "matchPattern": "(?i)^(openai/)?(gpt-5.1)$",
+ "startDate": "2025-11-14T08:57:23.481Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00000125,
+ "input_cached_tokens": 1.25e-7,
+ "output": 0.00001,
+ "input_cache_read": 1.25e-7,
+ "output_reasoning_tokens": 0.00001,
+ "output_reasoning": 0.00001
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.1-2025-11-13",
+ "matchPattern": "(?i)^(openai/)?(gpt-5.1-2025-11-13)$",
+ "startDate": "2025-11-14T08:57:23.481Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00000125,
+ "input_cached_tokens": 1.25e-7,
+ "output": 0.00001,
+ "input_cache_read": 1.25e-7,
+ "output_reasoning_tokens": 0.00001,
+ "output_reasoning": 0.00001
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-opus-4-5-20251101",
+ "matchPattern": "(?i)^(anthropic/)?(claude-opus-4-5(-20251101)?|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-opus-4-5(-20251101)?-v1(:0)?|claude-opus-4-5(@20251101)?)$",
+ "startDate": "2025-11-24T20:53:27.571Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000005,
+ "input_tokens": 0.000005,
+ "output": 0.000025,
+ "output_tokens": 0.000025,
+ "cache_creation_input_tokens": 0.00000625,
+ "input_cache_creation": 0.00000625,
+ "input_cache_creation_5m": 0.00000625,
+ "input_cache_creation_1h": 0.00001,
+ "cache_read_input_tokens": 5e-7,
+ "input_cache_read": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-sonnet-4-6",
+ "matchPattern": "(?i)^(anthropic\\/)?(claude-sonnet-4-6|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-sonnet-4-6(-v1(:0)?)?|claude-sonnet-4-6)$",
+ "startDate": "2026-02-18T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000003,
+ "input_tokens": 0.000003,
+ "output": 0.000015,
+ "output_tokens": 0.000015,
+ "cache_creation_input_tokens": 0.00000375,
+ "input_cache_creation": 0.00000375,
+ "input_cache_creation_5m": 0.00000375,
+ "input_cache_creation_1h": 0.000006,
+ "cache_read_input_tokens": 3e-7,
+ "input_cache_read": 3e-7
+ }
+ },
+ {
+ "name": "Large Context",
+ "isDefault": false,
+ "priority": 1,
+ "conditions": [
{
- usageDetailPattern: "input",
- operator: "gt",
- value: 200000,
- },
+ "usageDetailPattern": "input",
+ "operator": "gt",
+ "value": 200000
+ }
],
- prices: {
- input: 0.000006,
- input_tokens: 0.000006,
- output: 0.0000225,
- output_tokens: 0.0000225,
- cache_creation_input_tokens: 0.0000075,
- input_cache_creation: 0.0000075,
- input_cache_creation_5m: 0.0000075,
- input_cache_creation_1h: 0.000012,
- cache_read_input_tokens: 6e-7,
- input_cache_read: 6e-7,
- },
- },
- ],
- },
- {
- modelName: "claude-opus-4-6",
- matchPattern:
- "(?i)^(anthropic/)?(claude-opus-4-6|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-opus-4-6-v1(:0)?|claude-opus-4-6)$",
- startDate: "2026-02-09T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000005,
- input_tokens: 0.000005,
- output: 0.000025,
- output_tokens: 0.000025,
- cache_creation_input_tokens: 0.00000625,
- input_cache_creation: 0.00000625,
- input_cache_creation_5m: 0.00000625,
- input_cache_creation_1h: 0.00001,
- cache_read_input_tokens: 5e-7,
- input_cache_read: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "gemini-2.5-pro",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-2.5-pro)$",
- startDate: "2025-11-26T13:27:53.545Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00000125,
- input_text: 0.00000125,
- input_modality_1: 0.00000125,
- prompt_token_count: 0.00000125,
- promptTokenCount: 0.00000125,
- input_cached_tokens: 1.25e-7,
- cached_content_token_count: 1.25e-7,
- output: 0.00001,
- output_text: 0.00001,
- output_modality_1: 0.00001,
- candidates_token_count: 0.00001,
- candidatesTokenCount: 0.00001,
- thoughtsTokenCount: 0.00001,
- thoughts_token_count: 0.00001,
- output_reasoning: 0.00001,
- },
- },
- {
- name: "Large Context",
- isDefault: false,
- priority: 1,
- conditions: [
+ "prices": {
+ "input": 0.000006,
+ "input_tokens": 0.000006,
+ "output": 0.0000225,
+ "output_tokens": 0.0000225,
+ "cache_creation_input_tokens": 0.0000075,
+ "input_cache_creation": 0.0000075,
+ "input_cache_creation_5m": 0.0000075,
+ "input_cache_creation_1h": 0.000012,
+ "cache_read_input_tokens": 6e-7,
+ "input_cache_read": 6e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "claude-opus-4-6",
+ "matchPattern": "(?i)^(anthropic/)?(claude-opus-4-6|(eu\\.|us\\.|apac\\.|global\\.)?anthropic\\.claude-opus-4-6-v1(:0)?|claude-opus-4-6)$",
+ "startDate": "2026-02-09T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000005,
+ "input_tokens": 0.000005,
+ "output": 0.000025,
+ "output_tokens": 0.000025,
+ "cache_creation_input_tokens": 0.00000625,
+ "input_cache_creation": 0.00000625,
+ "input_cache_creation_5m": 0.00000625,
+ "input_cache_creation_1h": 0.00001,
+ "cache_read_input_tokens": 5e-7,
+ "input_cache_read": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-2.5-pro",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-2.5-pro)$",
+ "startDate": "2025-11-26T13:27:53.545Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00000125,
+ "input_text": 0.00000125,
+ "input_modality_1": 0.00000125,
+ "prompt_token_count": 0.00000125,
+ "promptTokenCount": 0.00000125,
+ "input_cached_tokens": 1.25e-7,
+ "cached_content_token_count": 1.25e-7,
+ "output": 0.00001,
+ "output_text": 0.00001,
+ "output_modality_1": 0.00001,
+ "candidates_token_count": 0.00001,
+ "candidatesTokenCount": 0.00001,
+ "thoughtsTokenCount": 0.00001,
+ "thoughts_token_count": 0.00001,
+ "output_reasoning": 0.00001
+ }
+ },
+ {
+ "name": "Large Context",
+ "isDefault": false,
+ "priority": 1,
+ "conditions": [
{
- usageDetailPattern: "(input|prompt|cached)",
- operator: "gt",
- value: 200000,
- },
+ "usageDetailPattern": "(input|prompt|cached)",
+ "operator": "gt",
+ "value": 200000
+ }
],
- prices: {
- input: 0.0000025,
- input_text: 0.0000025,
- input_modality_1: 0.0000025,
- prompt_token_count: 0.0000025,
- promptTokenCount: 0.0000025,
- input_cached_tokens: 2.5e-7,
- cached_content_token_count: 2.5e-7,
- output: 0.000015,
- output_text: 0.000015,
- output_modality_1: 0.000015,
- candidates_token_count: 0.000015,
- candidatesTokenCount: 0.000015,
- thoughtsTokenCount: 0.000015,
- thoughts_token_count: 0.000015,
- output_reasoning: 0.000015,
- },
- },
- ],
- },
- {
- modelName: "gemini-3-pro-preview",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-3-pro-preview)$",
- startDate: "2025-11-26T13:27:53.545Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000002,
- input_text: 0.000002,
- input_modality_1: 0.000002,
- prompt_token_count: 0.000002,
- promptTokenCount: 0.000002,
- input_cached_tokens: 2e-7,
- cached_content_token_count: 2e-7,
- output: 0.000012,
- output_text: 0.000012,
- output_modality_1: 0.000012,
- candidates_token_count: 0.000012,
- candidatesTokenCount: 0.000012,
- thoughtsTokenCount: 0.000012,
- thoughts_token_count: 0.000012,
- output_reasoning: 0.000012,
- },
- },
- {
- name: "Large Context",
- isDefault: false,
- priority: 1,
- conditions: [
+ "prices": {
+ "input": 0.0000025,
+ "input_text": 0.0000025,
+ "input_modality_1": 0.0000025,
+ "prompt_token_count": 0.0000025,
+ "promptTokenCount": 0.0000025,
+ "input_cached_tokens": 2.5e-7,
+ "cached_content_token_count": 2.5e-7,
+ "output": 0.000015,
+ "output_text": 0.000015,
+ "output_modality_1": 0.000015,
+ "candidates_token_count": 0.000015,
+ "candidatesTokenCount": 0.000015,
+ "thoughtsTokenCount": 0.000015,
+ "thoughts_token_count": 0.000015,
+ "output_reasoning": 0.000015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-3-pro-preview",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-3-pro-preview)$",
+ "startDate": "2025-11-26T13:27:53.545Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000002,
+ "input_text": 0.000002,
+ "input_modality_1": 0.000002,
+ "prompt_token_count": 0.000002,
+ "promptTokenCount": 0.000002,
+ "input_cached_tokens": 2e-7,
+ "cached_content_token_count": 2e-7,
+ "output": 0.000012,
+ "output_text": 0.000012,
+ "output_modality_1": 0.000012,
+ "candidates_token_count": 0.000012,
+ "candidatesTokenCount": 0.000012,
+ "thoughtsTokenCount": 0.000012,
+ "thoughts_token_count": 0.000012,
+ "output_reasoning": 0.000012
+ }
+ },
+ {
+ "name": "Large Context",
+ "isDefault": false,
+ "priority": 1,
+ "conditions": [
{
- usageDetailPattern: "(input|prompt|cached)",
- operator: "gt",
- value: 200000,
- },
+ "usageDetailPattern": "(input|prompt|cached)",
+ "operator": "gt",
+ "value": 200000
+ }
],
- prices: {
- input: 0.000004,
- input_text: 0.000004,
- input_modality_1: 0.000004,
- prompt_token_count: 0.000004,
- promptTokenCount: 0.000004,
- input_cached_tokens: 4e-7,
- cached_content_token_count: 4e-7,
- output: 0.000018,
- output_text: 0.000018,
- output_modality_1: 0.000018,
- candidates_token_count: 0.000018,
- candidatesTokenCount: 0.000018,
- thoughtsTokenCount: 0.000018,
- thoughts_token_count: 0.000018,
- output_reasoning: 0.000018,
- },
- },
- ],
- },
- {
- modelName: "gemini-3.1-pro-preview",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-3.1-pro-preview(-customtools)?)$",
- startDate: "2026-02-19T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000002,
- input_modality_1: 0.000002,
- input_text: 0.000002,
- prompt_token_count: 0.000002,
- promptTokenCount: 0.000002,
- input_cached_tokens: 2e-7,
- cached_content_token_count: 2e-7,
- output: 0.000012,
- output_text: 0.000012,
- output_modality_1: 0.000012,
- candidates_token_count: 0.000012,
- candidatesTokenCount: 0.000012,
- thoughtsTokenCount: 0.000012,
- thoughts_token_count: 0.000012,
- output_reasoning: 0.000012,
- },
- },
- {
- name: "Large Context",
- isDefault: false,
- priority: 1,
- conditions: [
+ "prices": {
+ "input": 0.000004,
+ "input_text": 0.000004,
+ "input_modality_1": 0.000004,
+ "prompt_token_count": 0.000004,
+ "promptTokenCount": 0.000004,
+ "input_cached_tokens": 4e-7,
+ "cached_content_token_count": 4e-7,
+ "output": 0.000018,
+ "output_text": 0.000018,
+ "output_modality_1": 0.000018,
+ "candidates_token_count": 0.000018,
+ "candidatesTokenCount": 0.000018,
+ "thoughtsTokenCount": 0.000018,
+ "thoughts_token_count": 0.000018,
+ "output_reasoning": 0.000018
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-3.1-pro-preview",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-3.1-pro-preview(-customtools)?)$",
+ "startDate": "2026-02-19T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000002,
+ "input_modality_1": 0.000002,
+ "input_text": 0.000002,
+ "prompt_token_count": 0.000002,
+ "promptTokenCount": 0.000002,
+ "input_cached_tokens": 2e-7,
+ "cached_content_token_count": 2e-7,
+ "output": 0.000012,
+ "output_text": 0.000012,
+ "output_modality_1": 0.000012,
+ "candidates_token_count": 0.000012,
+ "candidatesTokenCount": 0.000012,
+ "thoughtsTokenCount": 0.000012,
+ "thoughts_token_count": 0.000012,
+ "output_reasoning": 0.000012
+ }
+ },
+ {
+ "name": "Large Context",
+ "isDefault": false,
+ "priority": 1,
+ "conditions": [
{
- usageDetailPattern: "(input|prompt|cached)",
- operator: "gt",
- value: 200000,
- },
+ "usageDetailPattern": "(input|prompt|cached)",
+ "operator": "gt",
+ "value": 200000
+ }
],
- prices: {
- input: 0.000004,
- input_modality_1: 0.000004,
- input_text: 0.000004,
- prompt_token_count: 0.000004,
- promptTokenCount: 0.000004,
- input_cached_tokens: 4e-7,
- cached_content_token_count: 4e-7,
- output: 0.000018,
- output_text: 0.000018,
- output_modality_1: 0.000018,
- candidates_token_count: 0.000018,
- candidatesTokenCount: 0.000018,
- thoughtsTokenCount: 0.000018,
- thoughts_token_count: 0.000018,
- output_reasoning: 0.000018,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.2",
- matchPattern: "(?i)^(openai/)?(gpt-5.2)$",
- startDate: "2025-12-12T09:00:06.513Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00000175,
- input_cached_tokens: 1.75e-7,
- input_cache_read: 1.75e-7,
- output: 0.000014,
- output_reasoning_tokens: 0.000014,
- output_reasoning: 0.000014,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.2-2025-12-11",
- matchPattern: "(?i)^(openai/)?(gpt-5.2-2025-12-11)$",
- startDate: "2025-12-12T09:00:06.513Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00000175,
- input_cached_tokens: 1.75e-7,
- input_cache_read: 1.75e-7,
- output: 0.000014,
- output_reasoning_tokens: 0.000014,
- output_reasoning: 0.000014,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.2-pro",
- matchPattern: "(?i)^(openai/)?(gpt-5.2-pro)$",
- startDate: "2025-12-12T09:00:06.513Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000021,
- output: 0.000168,
- output_reasoning_tokens: 0.000168,
- output_reasoning: 0.000168,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.2-pro-2025-12-11",
- matchPattern: "(?i)^(openai/)?(gpt-5.2-pro-2025-12-11)$",
- startDate: "2025-12-12T09:00:06.513Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.000021,
- output: 0.000168,
- output_reasoning_tokens: 0.000168,
- output_reasoning: 0.000168,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.4",
- matchPattern: "(?i)^(openai/)?(gpt-5.4)$",
- startDate: "2026-03-05T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000025,
- input_cached_tokens: 2.5e-7,
- input_cache_read: 2.5e-7,
- output: 0.000015,
- output_reasoning_tokens: 0.000015,
- output_reasoning: 0.000015,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.4-pro",
- matchPattern: "(?i)^(openai/)?(gpt-5.4-pro)$",
- startDate: "2026-03-05T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00003,
- output: 0.00018,
- output_reasoning_tokens: 0.00018,
- output_reasoning: 0.00018,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.4-2026-03-05",
- matchPattern: "(?i)^(openai/)?(gpt-5.4-2026-03-05)$",
- startDate: "2026-03-05T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.0000025,
- input_cached_tokens: 2.5e-7,
- input_cache_read: 2.5e-7,
- output: 0.000015,
- output_reasoning_tokens: 0.000015,
- output_reasoning: 0.000015,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.4-pro-2026-03-05",
- matchPattern: "(?i)^(openai/)?(gpt-5.4-pro-2026-03-05)$",
- startDate: "2026-03-05T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 0.00003,
- output: 0.00018,
- output_reasoning_tokens: 0.00018,
- output_reasoning: 0.00018,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.4-mini",
- matchPattern: "(?i)^(openai\\/)?(gpt-5.4-mini)$",
- startDate: "2026-03-18T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 7.5e-7,
- input_cached_tokens: 7.5e-8,
- input_cache_read: 7.5e-8,
- output: 0.0000045,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.4-mini-2026-03-17",
- matchPattern: "(?i)^(openai\\/)?(gpt-5.4-mini-2026-03-17)$",
- startDate: "2026-03-18T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 7.5e-7,
- input_cached_tokens: 7.5e-8,
- input_cache_read: 7.5e-8,
- output: 0.0000045,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.4-nano",
- matchPattern: "(?i)^(openai\\/)?(gpt-5.4-nano)$",
- startDate: "2026-03-18T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2e-7,
- input_cached_tokens: 2e-8,
- input_cache_read: 2e-8,
- output: 0.00000125,
- },
- },
- ],
- },
- {
- modelName: "gpt-5.4-nano-2026-03-17",
- matchPattern: "(?i)^(openai\\/)?(gpt-5.4-nano-2026-03-17)$",
- startDate: "2026-03-18T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2e-7,
- input_cached_tokens: 2e-8,
- input_cache_read: 2e-8,
- output: 0.00000125,
- },
- },
- ],
- },
- {
- modelName: "gemini-3-flash-preview",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-3-flash-preview)$",
- startDate: "2025-12-21T12:01:42.282Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 5e-7,
- input_text: 5e-7,
- input_modality_1: 5e-7,
- prompt_token_count: 5e-7,
- promptTokenCount: 5e-7,
- input_cached_tokens: 5e-8,
- cached_content_token_count: 5e-8,
- output: 0.000003,
- output_text: 0.000003,
- output_modality_1: 0.000003,
- candidates_token_count: 0.000003,
- candidatesTokenCount: 0.000003,
- thoughtsTokenCount: 0.000003,
- thoughts_token_count: 0.000003,
- output_reasoning: 0.000003,
- },
- },
- ],
- },
- {
- modelName: "gemini-3.1-flash-lite-preview",
- matchPattern: "(?i)^(google(ai)?/)?(gemini-3.1-flash-lite-preview)$",
- startDate: "2026-03-03T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input: 2.5e-7,
- input_modality_1: 2.5e-7,
- input_text: 2.5e-7,
- prompt_token_count: 2.5e-7,
- promptTokenCount: 2.5e-7,
- input_cached_tokens: 2.5e-8,
- cached_content_token_count: 2.5e-8,
- output: 0.0000015,
- output_text: 0.0000015,
- output_modality_1: 0.0000015,
- candidates_token_count: 0.0000015,
- candidatesTokenCount: 0.0000015,
- thoughtsTokenCount: 0.0000015,
- thoughts_token_count: 0.0000015,
- output_reasoning: 0.0000015,
- input_audio_tokens: 5e-7,
- },
- },
- ],
- },
- {
- modelName: "gemini-live-2.5-flash-native-audio",
- matchPattern: "(?i)^(google/)?(gemini-live-2.5-flash-native-audio)$",
- startDate: "2026-03-16T00:00:00.000Z",
- pricingTiers: [
- {
- name: "Standard",
- isDefault: true,
- priority: 0,
- conditions: [],
- prices: {
- input_text: 5e-7,
- input_audio: 0.000003,
- input_image: 0.000003,
- output_text: 0.000002,
- output_audio: 0.000012,
- },
- },
- ],
- },
+ "prices": {
+ "input": 0.000004,
+ "input_modality_1": 0.000004,
+ "input_text": 0.000004,
+ "prompt_token_count": 0.000004,
+ "promptTokenCount": 0.000004,
+ "input_cached_tokens": 4e-7,
+ "cached_content_token_count": 4e-7,
+ "output": 0.000018,
+ "output_text": 0.000018,
+ "output_modality_1": 0.000018,
+ "candidates_token_count": 0.000018,
+ "candidatesTokenCount": 0.000018,
+ "thoughtsTokenCount": 0.000018,
+ "thoughts_token_count": 0.000018,
+ "output_reasoning": 0.000018
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.2",
+ "matchPattern": "(?i)^(openai/)?(gpt-5.2)$",
+ "startDate": "2025-12-12T09:00:06.513Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00000175,
+ "input_cached_tokens": 1.75e-7,
+ "input_cache_read": 1.75e-7,
+ "output": 0.000014,
+ "output_reasoning_tokens": 0.000014,
+ "output_reasoning": 0.000014
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.2-2025-12-11",
+ "matchPattern": "(?i)^(openai/)?(gpt-5.2-2025-12-11)$",
+ "startDate": "2025-12-12T09:00:06.513Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00000175,
+ "input_cached_tokens": 1.75e-7,
+ "input_cache_read": 1.75e-7,
+ "output": 0.000014,
+ "output_reasoning_tokens": 0.000014,
+ "output_reasoning": 0.000014
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.2-pro",
+ "matchPattern": "(?i)^(openai/)?(gpt-5.2-pro)$",
+ "startDate": "2025-12-12T09:00:06.513Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000021,
+ "output": 0.000168,
+ "output_reasoning_tokens": 0.000168,
+ "output_reasoning": 0.000168
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.2-pro-2025-12-11",
+ "matchPattern": "(?i)^(openai/)?(gpt-5.2-pro-2025-12-11)$",
+ "startDate": "2025-12-12T09:00:06.513Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.000021,
+ "output": 0.000168,
+ "output_reasoning_tokens": 0.000168,
+ "output_reasoning": 0.000168
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.4",
+ "matchPattern": "(?i)^(openai/)?(gpt-5.4)$",
+ "startDate": "2026-03-05T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000025,
+ "input_cached_tokens": 2.5e-7,
+ "input_cache_read": 2.5e-7,
+ "output": 0.000015,
+ "output_reasoning_tokens": 0.000015,
+ "output_reasoning": 0.000015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.4-pro",
+ "matchPattern": "(?i)^(openai/)?(gpt-5.4-pro)$",
+ "startDate": "2026-03-05T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00003,
+ "output": 0.00018,
+ "output_reasoning_tokens": 0.00018,
+ "output_reasoning": 0.00018
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.4-2026-03-05",
+ "matchPattern": "(?i)^(openai/)?(gpt-5.4-2026-03-05)$",
+ "startDate": "2026-03-05T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.0000025,
+ "input_cached_tokens": 2.5e-7,
+ "input_cache_read": 2.5e-7,
+ "output": 0.000015,
+ "output_reasoning_tokens": 0.000015,
+ "output_reasoning": 0.000015
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.4-pro-2026-03-05",
+ "matchPattern": "(?i)^(openai/)?(gpt-5.4-pro-2026-03-05)$",
+ "startDate": "2026-03-05T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 0.00003,
+ "output": 0.00018,
+ "output_reasoning_tokens": 0.00018,
+ "output_reasoning": 0.00018
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.4-mini",
+ "matchPattern": "(?i)^(openai\\/)?(gpt-5.4-mini)$",
+ "startDate": "2026-03-18T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 7.5e-7,
+ "input_cached_tokens": 7.5e-8,
+ "input_cache_read": 7.5e-8,
+ "output": 0.0000045
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.4-mini-2026-03-17",
+ "matchPattern": "(?i)^(openai\\/)?(gpt-5.4-mini-2026-03-17)$",
+ "startDate": "2026-03-18T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 7.5e-7,
+ "input_cached_tokens": 7.5e-8,
+ "input_cache_read": 7.5e-8,
+ "output": 0.0000045
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.4-nano",
+ "matchPattern": "(?i)^(openai\\/)?(gpt-5.4-nano)$",
+ "startDate": "2026-03-18T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2e-7,
+ "input_cached_tokens": 2e-8,
+ "input_cache_read": 2e-8,
+ "output": 0.00000125
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gpt-5.4-nano-2026-03-17",
+ "matchPattern": "(?i)^(openai\\/)?(gpt-5.4-nano-2026-03-17)$",
+ "startDate": "2026-03-18T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2e-7,
+ "input_cached_tokens": 2e-8,
+ "input_cache_read": 2e-8,
+ "output": 0.00000125
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-3-flash-preview",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-3-flash-preview)$",
+ "startDate": "2025-12-21T12:01:42.282Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 5e-7,
+ "input_text": 5e-7,
+ "input_modality_1": 5e-7,
+ "prompt_token_count": 5e-7,
+ "promptTokenCount": 5e-7,
+ "input_cached_tokens": 5e-8,
+ "cached_content_token_count": 5e-8,
+ "output": 0.000003,
+ "output_text": 0.000003,
+ "output_modality_1": 0.000003,
+ "candidates_token_count": 0.000003,
+ "candidatesTokenCount": 0.000003,
+ "thoughtsTokenCount": 0.000003,
+ "thoughts_token_count": 0.000003,
+ "output_reasoning": 0.000003
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-3.1-flash-lite-preview",
+ "matchPattern": "(?i)^(google(ai)?/)?(gemini-3.1-flash-lite-preview)$",
+ "startDate": "2026-03-03T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input": 2.5e-7,
+ "input_modality_1": 2.5e-7,
+ "input_text": 2.5e-7,
+ "prompt_token_count": 2.5e-7,
+ "promptTokenCount": 2.5e-7,
+ "input_cached_tokens": 2.5e-8,
+ "cached_content_token_count": 2.5e-8,
+ "output": 0.0000015,
+ "output_text": 0.0000015,
+ "output_modality_1": 0.0000015,
+ "candidates_token_count": 0.0000015,
+ "candidatesTokenCount": 0.0000015,
+ "thoughtsTokenCount": 0.0000015,
+ "thoughts_token_count": 0.0000015,
+ "output_reasoning": 0.0000015,
+ "input_audio_tokens": 5e-7
+ }
+ }
+ ]
+ },
+ {
+ "modelName": "gemini-live-2.5-flash-native-audio",
+ "matchPattern": "(?i)^(google/)?(gemini-live-2.5-flash-native-audio)$",
+ "startDate": "2026-03-16T00:00:00.000Z",
+ "pricingTiers": [
+ {
+ "name": "Standard",
+ "isDefault": true,
+ "priority": 0,
+ "conditions": [],
+ "prices": {
+ "input_text": 5e-7,
+ "input_audio": 0.000003,
+ "input_image": 0.000003,
+ "output_text": 0.000002,
+ "output_audio": 0.000012
+ }
+ }
+ ]
+ }
];
diff --git a/internal-packages/llm-model-catalog/src/modelCatalog.ts b/internal-packages/llm-model-catalog/src/modelCatalog.ts
index d90eb1a992b..71ae921c3e7 100644
--- a/internal-packages/llm-model-catalog/src/modelCatalog.ts
+++ b/internal-packages/llm-model-catalog/src/modelCatalog.ts
@@ -5,564 +5,677 @@ import type { ModelCatalogEntry } from "./types.js";
export const modelCatalog: Record = {
"chatgpt-4o-latest": {
- provider: "openai",
- description:
- "OpenAI's flagship multimodal model optimized for speed and cost, capable of processing text, images, and audio with strong performance across reasoning, coding, and creative tasks.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: [
+ "provider": "openai",
+ "description": "OpenAI's flagship multimodal model optimized for speed and cost, capable of processing text, images, and audio with strong performance across reasoning, coding, and creative tasks.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"audio_input",
"audio_output",
- "fine_tunable",
+ "fine_tunable"
],
- releaseDate: "2024-05-13",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T10:55:46.469Z",
- baseModelName: "chatgpt-4o",
+ "releaseDate": "2024-05-13",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T10:55:46.469Z",
+ "baseModelName": "chatgpt-4o"
},
"claude-1.1": {
- provider: "anthropic",
- description:
- "An early-generation Claude model from Anthropic, offering basic conversational and text completion capabilities. It was quickly superseded by Claude 1.2, 1.3, and the Claude 2 family.",
- contextWindow: 9000,
- maxOutputTokens: 8191,
- capabilities: ["streaming"],
- releaseDate: "2023-03-14",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: null,
- resolvedAt: "2026-03-24T10:55:47.906Z",
- baseModelName: null,
+ "provider": "anthropic",
+ "description": "An early-generation Claude model from Anthropic, offering basic conversational and text completion capabilities. It was quickly superseded by Claude 1.2, 1.3, and the Claude 2 family.",
+ "contextWindow": 9000,
+ "maxOutputTokens": 8191,
+ "capabilities": [
+ "streaming"
+ ],
+ "releaseDate": "2023-03-14",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": null,
+ "resolvedAt": "2026-03-24T10:55:47.906Z",
+ "baseModelName": null
},
"claude-1.2": {
- provider: "anthropic",
- description:
- "An early-generation Anthropic model, part of the original Claude 1.x family. It offered improved performance over Claude 1.0 but was quickly superseded by Claude 1.3 and later model families.",
- contextWindow: 9000,
- maxOutputTokens: 8191,
- capabilities: ["streaming"],
- releaseDate: null,
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: null,
- resolvedAt: "2026-03-24T10:55:46.760Z",
- baseModelName: null,
+ "provider": "anthropic",
+ "description": "An early-generation Anthropic model, part of the original Claude 1.x family. It offered improved performance over Claude 1.0 but was quickly superseded by Claude 1.3 and later model families.",
+ "contextWindow": 9000,
+ "maxOutputTokens": 8191,
+ "capabilities": [
+ "streaming"
+ ],
+ "releaseDate": null,
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": null,
+ "resolvedAt": "2026-03-24T10:55:46.760Z",
+ "baseModelName": null
},
"claude-1.3": {
- provider: "anthropic",
- description:
- "Early-generation Claude model from Anthropic, offering improved performance over Claude 1.0-1.2 in reasoning and instruction-following tasks.",
- contextWindow: 100000,
- maxOutputTokens: null,
- capabilities: ["streaming"],
- releaseDate: "2023-03-14",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: null,
- resolvedAt: "2026-03-24T10:55:46.227Z",
- baseModelName: null,
+ "provider": "anthropic",
+ "description": "Early-generation Claude model from Anthropic, offering improved performance over Claude 1.0-1.2 in reasoning and instruction-following tasks.",
+ "contextWindow": 100000,
+ "maxOutputTokens": null,
+ "capabilities": [
+ "streaming"
+ ],
+ "releaseDate": "2023-03-14",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": null,
+ "resolvedAt": "2026-03-24T10:55:46.227Z",
+ "baseModelName": null
},
"claude-2.0": {
- provider: "anthropic",
- description:
- "Anthropic's second-generation large language model, offering improved performance over Claude 1.x with longer context support. Succeeded by Claude 2.1 and later the Claude 3 family.",
- contextWindow: 100000,
- maxOutputTokens: 4096,
- capabilities: ["streaming"],
- releaseDate: "2023-07-11",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: "2023-02-01",
- resolvedAt: "2026-03-24T10:55:45.922Z",
- baseModelName: null,
+ "provider": "anthropic",
+ "description": "Anthropic's second-generation large language model, offering improved performance over Claude 1.x with longer context support. Succeeded by Claude 2.1 and later the Claude 3 family.",
+ "contextWindow": 100000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "streaming"
+ ],
+ "releaseDate": "2023-07-11",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-02-01",
+ "resolvedAt": "2026-03-24T10:55:45.922Z",
+ "baseModelName": null
},
"claude-2.1": {
- provider: "anthropic",
- description:
- "Anthropic's Claude 2.1 model featuring a 200K context window, reduced hallucination rates compared to Claude 2.0, and improved accuracy on long document comprehension.",
- contextWindow: 200000,
- maxOutputTokens: 4096,
- capabilities: ["streaming", "tool_use"],
- releaseDate: "2023-11-21",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: "2023-01-01",
- resolvedAt: "2026-03-24T10:56:22.743Z",
- baseModelName: null,
+ "provider": "anthropic",
+ "description": "Anthropic's Claude 2.1 model featuring a 200K context window, reduced hallucination rates compared to Claude 2.0, and improved accuracy on long document comprehension.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "streaming",
+ "tool_use"
+ ],
+ "releaseDate": "2023-11-21",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-01-01",
+ "resolvedAt": "2026-03-24T10:56:22.743Z",
+ "baseModelName": null
},
"claude-3-5-haiku-20241022": {
- provider: "anthropic",
- description:
- "Anthropic's fastest and most cost-effective model in the Claude 3.5 family, optimized for speed and efficiency while maintaining strong performance across common tasks.",
- contextWindow: 200000,
- maxOutputTokens: 8192,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-10-22",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-07-01",
- resolvedAt: "2026-03-24T10:56:25.724Z",
- baseModelName: "claude-3-5-haiku",
+ "provider": "anthropic",
+ "description": "Anthropic's fastest and most cost-effective model in the Claude 3.5 family, optimized for speed and efficiency while maintaining strong performance across common tasks.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-10-22",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-07-01",
+ "resolvedAt": "2026-03-24T10:56:25.724Z",
+ "baseModelName": "claude-3-5-haiku"
},
"claude-3-5-sonnet-20240620": {
- provider: "anthropic",
- description:
- "Anthropic's Claude 3.5 Sonnet is a mid-tier model balancing intelligence and speed, excelling at coding, analysis, and vision tasks while being faster and cheaper than Opus.",
- contextWindow: 200000,
- maxOutputTokens: 8192,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-06-20",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-04-01",
- resolvedAt: "2026-03-24T10:56:35.401Z",
- baseModelName: "claude-3-5-sonnet",
+ "provider": "anthropic",
+ "description": "Anthropic's Claude 3.5 Sonnet is a mid-tier model balancing intelligence and speed, excelling at coding, analysis, and vision tasks while being faster and cheaper than Opus.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-06-20",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-04-01",
+ "resolvedAt": "2026-03-24T10:56:35.401Z",
+ "baseModelName": "claude-3-5-sonnet"
},
"claude-3-haiku-20240307": {
- provider: "anthropic",
- description:
- "Anthropic's fastest and most compact Claude 3 model, optimized for speed and cost-efficiency while maintaining strong performance on everyday tasks.",
- contextWindow: 200000,
- maxOutputTokens: 4096,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-03-13",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-08-01",
- resolvedAt: "2026-03-24T10:56:25.288Z",
- baseModelName: "claude-3-haiku",
+ "provider": "anthropic",
+ "description": "Anthropic's fastest and most compact Claude 3 model, optimized for speed and cost-efficiency while maintaining strong performance on everyday tasks.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-03-13",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-08-01",
+ "resolvedAt": "2026-03-24T10:56:25.288Z",
+ "baseModelName": "claude-3-haiku"
},
"claude-3-opus-20240229": {
- provider: "anthropic",
- description:
- "Anthropic's most capable model in the Claude 3 family, excelling at complex analysis, nuanced content generation, and advanced reasoning tasks.",
- contextWindow: 200000,
- maxOutputTokens: 4096,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-03-04",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-08-01",
- resolvedAt: "2026-03-24T10:56:26.008Z",
- baseModelName: "claude-3-opus",
+ "provider": "anthropic",
+ "description": "Anthropic's most capable model in the Claude 3 family, excelling at complex analysis, nuanced content generation, and advanced reasoning tasks.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-03-04",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-08-01",
+ "resolvedAt": "2026-03-24T10:56:26.008Z",
+ "baseModelName": "claude-3-opus"
},
"claude-3-sonnet-20240229": {
- provider: "anthropic",
- description:
- "Mid-tier model in Anthropic's Claude 3 family, balancing performance and speed for a wide range of tasks including analysis, coding, and content generation.",
- contextWindow: 200000,
- maxOutputTokens: 4096,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-03-04",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-02-01",
- resolvedAt: "2026-03-24T10:56:59.532Z",
- baseModelName: "claude-3-sonnet",
+ "provider": "anthropic",
+ "description": "Mid-tier model in Anthropic's Claude 3 family, balancing performance and speed for a wide range of tasks including analysis, coding, and content generation.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-03-04",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-02-01",
+ "resolvedAt": "2026-03-24T10:56:59.532Z",
+ "baseModelName": "claude-3-sonnet"
},
"claude-3.5-haiku-latest": {
- provider: "anthropic",
- description:
- "Anthropic's fastest and most cost-effective model in the Claude 3.5 family, optimized for speed and efficiency while maintaining strong performance across a wide range of tasks.",
- contextWindow: 200000,
- maxOutputTokens: 8192,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-10-29",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-07-01",
- resolvedAt: "2026-03-24T10:57:04.392Z",
- baseModelName: "claude-3.5-haiku",
+ "provider": "anthropic",
+ "description": "Anthropic's fastest and most cost-effective model in the Claude 3.5 family, optimized for speed and efficiency while maintaining strong performance across a wide range of tasks.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-10-29",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-07-01",
+ "resolvedAt": "2026-03-24T10:57:04.392Z",
+ "baseModelName": "claude-3.5-haiku"
},
"claude-3.5-sonnet-20241022": {
- provider: "anthropic",
- description:
- "Anthropic's mid-tier model offering strong reasoning, coding, and analysis capabilities at a balance of speed and intelligence, positioned between Haiku and Opus in the Claude 3.5 family.",
- contextWindow: 200000,
- maxOutputTokens: 8192,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-06-20",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-04-01",
- resolvedAt: "2026-03-24T10:57:13.346Z",
- baseModelName: "claude-3.5-sonnet",
+ "provider": "anthropic",
+ "description": "Anthropic's mid-tier model offering strong reasoning, coding, and analysis capabilities at a balance of speed and intelligence, positioned between Haiku and Opus in the Claude 3.5 family.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-06-20",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-04-01",
+ "resolvedAt": "2026-03-24T10:57:13.346Z",
+ "baseModelName": "claude-3.5-sonnet"
},
"claude-3.5-sonnet-latest": {
- provider: "anthropic",
- description:
- "Anthropic's mid-tier model offering strong reasoning, coding, and analysis capabilities at a balance of speed and intelligence, positioned between Haiku and Opus in the Claude 3.5 family.",
- contextWindow: 200000,
- maxOutputTokens: 8192,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-06-20",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-04-01",
- resolvedAt: "2026-03-24T10:57:13.346Z",
- baseModelName: "claude-3.5-sonnet",
+ "provider": "anthropic",
+ "description": "Anthropic's mid-tier model offering strong reasoning, coding, and analysis capabilities at a balance of speed and intelligence, positioned between Haiku and Opus in the Claude 3.5 family.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-06-20",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-04-01",
+ "resolvedAt": "2026-03-24T10:57:13.346Z",
+ "baseModelName": "claude-3.5-sonnet"
},
"claude-3.7-sonnet-20250219": {
- provider: "anthropic",
- description:
- "Anthropic's Claude 3.7 Sonnet is a hybrid reasoning model that introduced extended thinking capabilities, offering strong performance on coding, math, and complex reasoning tasks.",
- contextWindow: 200000,
- maxOutputTokens: 16384,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-02-24",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-04-01",
- resolvedAt: "2026-03-24T10:57:12.967Z",
- baseModelName: "claude-3.7-sonnet",
+ "provider": "anthropic",
+ "description": "Anthropic's Claude 3.7 Sonnet is a hybrid reasoning model that introduced extended thinking capabilities, offering strong performance on coding, math, and complex reasoning tasks.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-02-24",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-04-01",
+ "resolvedAt": "2026-03-24T10:57:12.967Z",
+ "baseModelName": "claude-3.7-sonnet"
},
"claude-3.7-sonnet-latest": {
- provider: "anthropic",
- description:
- "Anthropic's Claude 3.7 Sonnet is a hybrid reasoning model that introduced extended thinking capabilities, offering strong performance on coding, math, and complex reasoning tasks.",
- contextWindow: 200000,
- maxOutputTokens: 16384,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-02-24",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-04-01",
- resolvedAt: "2026-03-24T10:57:12.967Z",
- baseModelName: "claude-3.7-sonnet",
+ "provider": "anthropic",
+ "description": "Anthropic's Claude 3.7 Sonnet is a hybrid reasoning model that introduced extended thinking capabilities, offering strong performance on coding, math, and complex reasoning tasks.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-02-24",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-04-01",
+ "resolvedAt": "2026-03-24T10:57:12.967Z",
+ "baseModelName": "claude-3.7-sonnet"
},
"claude-haiku-4-5-20251001": {
- provider: "anthropic",
- description:
- "Anthropic's fastest model with near-frontier intelligence, optimized for speed and cost efficiency while supporting extended thinking and vision.",
- contextWindow: 200000,
- maxOutputTokens: 64000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-10-01",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-07-01",
- resolvedAt: "2026-03-24T10:57:29.685Z",
- baseModelName: "claude-haiku-4-5",
+ "provider": "anthropic",
+ "description": "Anthropic's fastest model with near-frontier intelligence, optimized for speed and cost efficiency while supporting extended thinking and vision.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 64000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-10-01",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-07-01",
+ "resolvedAt": "2026-03-24T10:57:29.685Z",
+ "baseModelName": "claude-haiku-4-5"
},
"claude-instant-1": {
- provider: "anthropic",
- description:
- "Anthropic's fast and cost-effective model optimized for speed and efficiency, positioned as a lighter alternative to Claude 1.x for tasks requiring lower latency.",
- contextWindow: 100000,
- maxOutputTokens: 8191,
- capabilities: ["streaming"],
- releaseDate: "2023-03-14",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-01-06",
- knowledgeCutoff: "2023-01-01",
- resolvedAt: "2026-03-24T10:57:36.888Z",
- baseModelName: null,
+ "provider": "anthropic",
+ "description": "Anthropic's fast and cost-effective model optimized for speed and efficiency, positioned as a lighter alternative to Claude 1.x for tasks requiring lower latency.",
+ "contextWindow": 100000,
+ "maxOutputTokens": 8191,
+ "capabilities": [
+ "streaming"
+ ],
+ "releaseDate": "2023-03-14",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-01-06",
+ "knowledgeCutoff": "2023-01-01",
+ "resolvedAt": "2026-03-24T10:57:36.888Z",
+ "baseModelName": null
},
"claude-instant-1.2": {
- provider: "anthropic",
- description:
- "Anthropic's fast and cost-effective model, optimized for speed and efficiency while maintaining strong performance on conversational and text generation tasks.",
- contextWindow: 100000,
- maxOutputTokens: 8191,
- capabilities: ["streaming"],
- releaseDate: "2023-08-09",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: "2023-01-01",
- resolvedAt: "2026-03-24T10:57:41.865Z",
- baseModelName: null,
+ "provider": "anthropic",
+ "description": "Anthropic's fast and cost-effective model, optimized for speed and efficiency while maintaining strong performance on conversational and text generation tasks.",
+ "contextWindow": 100000,
+ "maxOutputTokens": 8191,
+ "capabilities": [
+ "streaming"
+ ],
+ "releaseDate": "2023-08-09",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-01-01",
+ "resolvedAt": "2026-03-24T10:57:41.865Z",
+ "baseModelName": null
},
"claude-opus-4-1-20250805": {
- provider: "anthropic",
- description:
- "Anthropic's hybrid reasoning model with strong software engineering and agentic capabilities, scoring 74.5% on SWE-bench Verified. Supports both rapid responses and step-by-step extended thinking.",
- contextWindow: 200000,
- maxOutputTokens: 32000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-08-05",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-03-01",
- resolvedAt: "2026-03-24T10:58:36.876Z",
- baseModelName: "claude-opus-4-1",
+ "provider": "anthropic",
+ "description": "Anthropic's hybrid reasoning model with strong software engineering and agentic capabilities, scoring 74.5% on SWE-bench Verified. Supports both rapid responses and step-by-step extended thinking.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 32000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-08-05",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-03-01",
+ "resolvedAt": "2026-03-24T10:58:36.876Z",
+ "baseModelName": "claude-opus-4-1"
},
"claude-opus-4-20250514": {
- provider: "anthropic",
- description:
- "Anthropic's flagship model from the Claude 4 family, excelling at complex coding tasks, long-running agent workflows, and deep reasoning with extended thinking support.",
- contextWindow: 200000,
- maxOutputTokens: 32000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-05-14",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-03-01",
- resolvedAt: "2026-03-24T10:58:47.518Z",
- baseModelName: "claude-opus-4",
+ "provider": "anthropic",
+ "description": "Anthropic's flagship model from the Claude 4 family, excelling at complex coding tasks, long-running agent workflows, and deep reasoning with extended thinking support.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 32000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-05-14",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-03-01",
+ "resolvedAt": "2026-03-24T10:58:47.518Z",
+ "baseModelName": "claude-opus-4"
},
"claude-opus-4-5-20251101": {
- provider: "anthropic",
- description:
- "Anthropic's flagship intelligence model released in November 2025, excelling at complex reasoning, vision, and extended thinking with the best performance in Anthropic's lineup before Opus 4.6.",
- contextWindow: 200000,
- maxOutputTokens: 64000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-11-01",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-03-01",
- resolvedAt: "2026-03-24T10:58:48.961Z",
- baseModelName: "claude-opus-4-5",
+ "provider": "anthropic",
+ "description": "Anthropic's flagship intelligence model released in November 2025, excelling at complex reasoning, vision, and extended thinking with the best performance in Anthropic's lineup before Opus 4.6.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 64000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-11-01",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-03-01",
+ "resolvedAt": "2026-03-24T10:58:48.961Z",
+ "baseModelName": "claude-opus-4-5"
},
"claude-opus-4-6": {
- provider: "anthropic",
- description:
- "Anthropic's most intelligent model, optimized for building agents and coding with exceptional reasoning capabilities and extended agentic task horizons.",
- contextWindow: 1000000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2026-02-05",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-05-01",
- resolvedAt: "2026-03-24T10:58:42.061Z",
- baseModelName: null,
+ "provider": "anthropic",
+ "description": "Anthropic's most intelligent model, optimized for building agents and coding with exceptional reasoning capabilities and extended agentic task horizons.",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2026-02-05",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-05-01",
+ "resolvedAt": "2026-03-24T10:58:42.061Z",
+ "baseModelName": null
},
"claude-sonnet-4-20250514": {
- provider: "anthropic",
- description:
- "Anthropic's balanced Claude 4 model offering strong coding, reasoning, and multilingual performance at moderate cost. Now a legacy model superseded by Claude Sonnet 4.5 and 4.6.",
- contextWindow: 200000,
- maxOutputTokens: 64000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-05-14",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-03-01",
- resolvedAt: "2026-03-24T10:58:39.601Z",
- baseModelName: "claude-sonnet-4",
+ "provider": "anthropic",
+ "description": "Anthropic's balanced Claude 4 model offering strong coding, reasoning, and multilingual performance at moderate cost. Now a legacy model superseded by Claude Sonnet 4.5 and 4.6.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 64000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-05-14",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-03-01",
+ "resolvedAt": "2026-03-24T10:58:39.601Z",
+ "baseModelName": "claude-sonnet-4"
},
"claude-sonnet-4-5-20250929": {
- provider: "anthropic",
- description:
- "Anthropic's high-performance mid-tier model with strong coding, reasoning, and multi-step problem solving capabilities. Successor to Claude Sonnet 4, offering improved benchmarks at the same price point.",
- contextWindow: 200000,
- maxOutputTokens: 64000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-09-29",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-01-01",
- resolvedAt: "2026-03-24T10:59:54.426Z",
- baseModelName: "claude-sonnet-4-5",
+ "provider": "anthropic",
+ "description": "Anthropic's high-performance mid-tier model with strong coding, reasoning, and multi-step problem solving capabilities. Successor to Claude Sonnet 4, offering improved benchmarks at the same price point.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 64000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-09-29",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-01-01",
+ "resolvedAt": "2026-03-24T10:59:54.426Z",
+ "baseModelName": "claude-sonnet-4-5"
},
"claude-sonnet-4-6": {
- provider: "anthropic",
- description:
- "Anthropic's best combination of speed and intelligence, excelling at coding, agentic tasks, and computer use, with a 1M token context window and performance rivaling prior Opus-class models.",
- contextWindow: 1000000,
- maxOutputTokens: 64000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2026-02-17",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2026-01-01",
- resolvedAt: "2026-03-24T10:59:59.014Z",
- baseModelName: null,
+ "provider": "anthropic",
+ "description": "Anthropic's best combination of speed and intelligence, excelling at coding, agentic tasks, and computer use, with a 1M token context window and performance rivaling prior Opus-class models.",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 64000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2026-02-17",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2026-01-01",
+ "resolvedAt": "2026-03-24T10:59:59.014Z",
+ "baseModelName": null
},
"claude-sonnet-4-latest": {
- provider: "anthropic",
- description:
- "Anthropic's balanced Claude 4 model offering strong coding, reasoning, and multilingual performance at moderate cost. Now a legacy model superseded by Claude Sonnet 4.5 and 4.6.",
- contextWindow: 200000,
- maxOutputTokens: 64000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-05-14",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-03-01",
- resolvedAt: "2026-03-24T10:58:39.601Z",
- baseModelName: "claude-sonnet-4",
+ "provider": "anthropic",
+ "description": "Anthropic's balanced Claude 4 model offering strong coding, reasoning, and multilingual performance at moderate cost. Now a legacy model superseded by Claude Sonnet 4.5 and 4.6.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 64000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-05-14",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-03-01",
+ "resolvedAt": "2026-03-24T10:58:39.601Z",
+ "baseModelName": "claude-sonnet-4"
},
"gemini-1.0-pro": {
- provider: "google",
- description:
- "Google's first-generation Gemini Pro model, a mid-size multimodal model designed for text generation, reasoning, and chat applications. Succeeded by Gemini 1.5 Pro.",
- contextWindow: 32760,
- maxOutputTokens: 8192,
- capabilities: ["tool_use", "streaming", "json_mode"],
- releaseDate: "2023-12-13",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-02-15",
- knowledgeCutoff: "2023-04-01",
- resolvedAt: "2026-03-24T10:59:26.767Z",
- baseModelName: null,
+ "provider": "google",
+ "description": "Google's first-generation Gemini Pro model, a mid-size multimodal model designed for text generation, reasoning, and chat applications. Succeeded by Gemini 1.5 Pro.",
+ "contextWindow": 32760,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2023-12-13",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-02-15",
+ "knowledgeCutoff": "2023-04-01",
+ "resolvedAt": "2026-03-24T10:59:26.767Z",
+ "baseModelName": null
},
"gemini-1.0-pro-001": {
- provider: "google",
- description:
- "Google's first-generation Pro model optimized for text generation, reasoning, and multi-turn conversation tasks, part of the original Gemini 1.0 lineup.",
- contextWindow: 30720,
- maxOutputTokens: 2048,
- capabilities: ["tool_use", "streaming", "json_mode"],
- releaseDate: "2024-02-15",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-02-15",
- knowledgeCutoff: "2023-04-01",
- resolvedAt: "2026-03-24T10:59:27.391Z",
- baseModelName: null,
+ "provider": "google",
+ "description": "Google's first-generation Pro model optimized for text generation, reasoning, and multi-turn conversation tasks, part of the original Gemini 1.0 lineup.",
+ "contextWindow": 30720,
+ "maxOutputTokens": 2048,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-02-15",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-02-15",
+ "knowledgeCutoff": "2023-04-01",
+ "resolvedAt": "2026-03-24T10:59:27.391Z",
+ "baseModelName": null
},
"gemini-1.0-pro-latest": {
- provider: "google",
- description:
- "Google's first-generation Gemini Pro model, a mid-size multimodal model designed for text generation, reasoning, and chat applications. Succeeded by Gemini 1.5 Pro.",
- contextWindow: 32760,
- maxOutputTokens: 8192,
- capabilities: ["tool_use", "streaming", "json_mode"],
- releaseDate: "2023-12-13",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-02-15",
- knowledgeCutoff: "2023-04-01",
- resolvedAt: "2026-03-24T10:59:26.767Z",
- baseModelName: "gemini-1.0-pro",
+ "provider": "google",
+ "description": "Google's first-generation Gemini Pro model, a mid-size multimodal model designed for text generation, reasoning, and chat applications. Succeeded by Gemini 1.5 Pro.",
+ "contextWindow": 32760,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2023-12-13",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-02-15",
+ "knowledgeCutoff": "2023-04-01",
+ "resolvedAt": "2026-03-24T10:59:26.767Z",
+ "baseModelName": "gemini-1.0-pro"
},
"gemini-1.5-pro-latest": {
- provider: "google",
- description:
- "Google's mid-size multimodal model with a massive context window, strong at long-document understanding, code generation, and multi-turn conversation.",
- contextWindow: 2097152,
- maxOutputTokens: 8192,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "audio_input"],
- releaseDate: "2024-02-15",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2025-09-24",
- knowledgeCutoff: "2024-04-01",
- resolvedAt: "2026-03-24T10:59:25.463Z",
- baseModelName: "gemini-1.5-pro",
+ "provider": "google",
+ "description": "Google's mid-size multimodal model with a massive context window, strong at long-document understanding, code generation, and multi-turn conversation.",
+ "contextWindow": 2097152,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "audio_input"
+ ],
+ "releaseDate": "2024-02-15",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2025-09-24",
+ "knowledgeCutoff": "2024-04-01",
+ "resolvedAt": "2026-03-24T10:59:25.463Z",
+ "baseModelName": "gemini-1.5-pro"
},
"gemini-2.0-flash": {
- provider: "google",
- description:
- "Google's second-generation workhorse model optimized for speed, with native tool use, multimodal input (text, images, audio, video), and a 1M token context window.",
- contextWindow: 1048576,
- maxOutputTokens: 8192,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "code_execution", "audio_input"],
- releaseDate: "2025-02-05",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2026-06-01",
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:01:15.429Z",
- baseModelName: null,
+ "provider": "google",
+ "description": "Google's second-generation workhorse model optimized for speed, with native tool use, multimodal input (text, images, audio, video), and a 1M token context window.",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "code_execution",
+ "audio_input"
+ ],
+ "releaseDate": "2025-02-05",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2026-06-01",
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:01:15.429Z",
+ "baseModelName": null
},
"gemini-2.0-flash-001": {
- provider: "google",
- description:
- "Google's fast and efficient multimodal model that outperforms Gemini 1.5 Pro on key benchmarks at twice the speed, supporting text, image, audio, and video inputs with native tool use.",
- contextWindow: 1048576,
- maxOutputTokens: 8192,
- capabilities: [
+ "provider": "google",
+ "description": "Google's fast and efficient multimodal model that outperforms Gemini 1.5 Pro on key benchmarks at twice the speed, supporting text, image, audio, and video inputs with native tool use.",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 8192,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
@@ -570,1614 +683,1890 @@ export const modelCatalog: Record = {
"audio_input",
"image_generation",
"audio_output",
- "code_execution",
+ "code_execution"
],
- releaseDate: "2025-02-05",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2026-06-01",
- knowledgeCutoff: "2024-08-01",
- resolvedAt: "2026-03-24T11:01:04.084Z",
- baseModelName: null,
+ "releaseDate": "2025-02-05",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2026-06-01",
+ "knowledgeCutoff": "2024-08-01",
+ "resolvedAt": "2026-03-24T11:01:04.084Z",
+ "baseModelName": null
},
"gemini-2.0-flash-lite-preview": {
- provider: "google",
- description:
- "A lightweight, cost-efficient variant of Gemini 2.0 Flash optimized for low latency and high throughput, supporting multimodal input with text output.",
- contextWindow: 1048576,
- maxOutputTokens: 8192,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-02-05",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2026-06-01",
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:00:56.775Z",
- baseModelName: null,
+ "provider": "google",
+ "description": "A lightweight, cost-efficient variant of Gemini 2.0 Flash optimized for low latency and high throughput, supporting multimodal input with text output.",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-02-05",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2026-06-01",
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:00:56.775Z",
+ "baseModelName": null
},
"gemini-2.0-flash-lite-preview-02-05": {
- provider: "google",
- description:
- "Google's cost-optimized, low-latency model in the Gemini 2.0 family, designed for high-volume tasks like summarization, multimodal processing, and categorization.",
- contextWindow: 1048576,
- maxOutputTokens: 8192,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-02-05",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2025-12-09",
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:01:34.165Z",
- baseModelName: null,
+ "provider": "google",
+ "description": "Google's cost-optimized, low-latency model in the Gemini 2.0 family, designed for high-volume tasks like summarization, multimodal processing, and categorization.",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-02-05",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2025-12-09",
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:01:34.165Z",
+ "baseModelName": null
},
"gemini-2.5-flash": {
- provider: "google",
- description:
- "Google's best price-performance model optimized for low-latency, high-volume tasks requiring reasoning, with built-in thinking capabilities and multimodal input support.",
- contextWindow: 1048576,
- maxOutputTokens: 65536,
- capabilities: [
+ "provider": "google",
+ "description": "Google's best price-performance model optimized for low-latency, high-volume tasks requiring reasoning, with built-in thinking capabilities and multimodal input support.",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 65536,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"extended_thinking",
"code_execution",
- "audio_input",
+ "audio_input"
],
- releaseDate: "2025-06-01",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: "2025-01-01",
- resolvedAt: "2026-03-24T11:01:25.200Z",
- baseModelName: null,
+ "releaseDate": "2025-06-01",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-01-01",
+ "resolvedAt": "2026-03-24T11:01:25.200Z",
+ "baseModelName": null
},
"gemini-2.5-flash-lite": {
- provider: "google",
- description:
- "Google's most cost-efficient Gemini model, optimized for low-latency use cases with strong reasoning, multilingual, and long-context capabilities at minimal cost.",
- contextWindow: 1048576,
- maxOutputTokens: 65535,
- capabilities: [
+ "provider": "google",
+ "description": "Google's most cost-efficient Gemini model, optimized for low-latency use cases with strong reasoning, multilingual, and long-context capabilities at minimal cost.",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 65535,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"extended_thinking",
"code_execution",
- "audio_input",
+ "audio_input"
],
- releaseDate: "2025-07-22",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2026-07-22",
- knowledgeCutoff: "2025-01-01",
- resolvedAt: "2026-03-24T11:02:30.060Z",
- baseModelName: null,
+ "releaseDate": "2025-07-22",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2026-07-22",
+ "knowledgeCutoff": "2025-01-01",
+ "resolvedAt": "2026-03-24T11:02:30.060Z",
+ "baseModelName": null
},
"gemini-2.5-pro": {
- provider: "google",
- description:
- "Google's most advanced reasoning model with deep thinking capabilities, excelling at complex tasks like coding, math, and multimodal understanding across text, images, audio, and video.",
- contextWindow: 1048576,
- maxOutputTokens: 65535,
- capabilities: [
+ "provider": "google",
+ "description": "Google's most advanced reasoning model with deep thinking capabilities, excelling at complex tasks like coding, math, and multimodal understanding across text, images, audio, and video.",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 65535,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"extended_thinking",
"code_execution",
- "audio_input",
+ "audio_input"
],
- releaseDate: "2025-03-25",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2026-06-17",
- knowledgeCutoff: "2025-01-31",
- resolvedAt: "2026-03-24T11:02:25.573Z",
- baseModelName: null,
+ "releaseDate": "2025-03-25",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2026-06-17",
+ "knowledgeCutoff": "2025-01-31",
+ "resolvedAt": "2026-03-24T11:02:25.573Z",
+ "baseModelName": null
},
"gemini-3-flash-preview": {
- provider: "google",
- description:
- "Google's high-speed thinking model that matches Gemini 2.5 Pro performance at ~3x faster speed and lower cost, designed for agentic workflows, multi-turn chat, and coding assistance with configurable reasoning levels.",
- contextWindow: 1048576,
- maxOutputTokens: 65536,
- capabilities: [
+ "provider": "google",
+ "description": "Google's high-speed thinking model that matches Gemini 2.5 Pro performance at ~3x faster speed and lower cost, designed for agentic workflows, multi-turn chat, and coding assistance with configurable reasoning levels.",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 65536,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"extended_thinking",
"code_execution",
- "audio_input",
+ "audio_input"
],
- releaseDate: "2025-12-17",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-01-01",
- resolvedAt: "2026-03-24T11:02:13.388Z",
- baseModelName: null,
+ "releaseDate": "2025-12-17",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-01-01",
+ "resolvedAt": "2026-03-24T11:02:13.388Z",
+ "baseModelName": null
},
"gemini-3-pro-preview": {
- provider: "google",
- description:
- "Google's flagship reasoning and multimodal model with strong coding and agentic capabilities, now deprecated in favor of Gemini 3.1 Pro.",
- contextWindow: 1048576,
- maxOutputTokens: 65536,
- capabilities: [
+ "provider": "google",
+ "description": "Google's flagship reasoning and multimodal model with strong coding and agentic capabilities, now deprecated in favor of Gemini 3.1 Pro.",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 65536,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"extended_thinking",
"code_execution",
- "audio_input",
+ "audio_input"
],
- releaseDate: "2025-11-01",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2026-03-09",
- knowledgeCutoff: "2025-01-01",
- resolvedAt: "2026-03-24T11:02:29.313Z",
- baseModelName: null,
+ "releaseDate": "2025-11-01",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2026-03-09",
+ "knowledgeCutoff": "2025-01-01",
+ "resolvedAt": "2026-03-24T11:02:29.313Z",
+ "baseModelName": null
},
"gemini-3.1-flash-lite-preview": {
- provider: "google",
- description:
- "Google's most cost-efficient multimodal model in the Gemini 3 series, optimized for high-volume, low-latency tasks like translation, classification, and simple data extraction. Offers 2.5x faster time-to-first-token than Gemini 2.5 Flash.",
- contextWindow: 1048576,
- maxOutputTokens: 65536,
- capabilities: [
+ "provider": "google",
+ "description": "Google's most cost-efficient multimodal model in the Gemini 3 series, optimized for high-volume, low-latency tasks like translation, classification, and simple data extraction. Offers 2.5x faster time-to-first-token than Gemini 2.5 Flash.",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 65536,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"extended_thinking",
"code_execution",
- "audio_input",
+ "audio_input"
],
- releaseDate: "2026-03-03",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-01-01",
- resolvedAt: "2026-03-24T11:02:29.253Z",
- baseModelName: null,
+ "releaseDate": "2026-03-03",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-01-01",
+ "resolvedAt": "2026-03-24T11:02:29.253Z",
+ "baseModelName": null
},
"gemini-3.1-pro-preview": {
- provider: "google",
- description:
- "Google's most advanced reasoning model in the Gemini 3.1 family, excelling at complex problem-solving across text, audio, images, video, and code with a 1M token context window and extended thinking capabilities.",
- contextWindow: 1048576,
- maxOutputTokens: 65536,
- capabilities: [
+ "provider": "google",
+ "description": "Google's most advanced reasoning model in the Gemini 3.1 family, excelling at complex problem-solving across text, audio, images, video, and code with a 1M token context window and extended thinking capabilities.",
+ "contextWindow": 1048576,
+ "maxOutputTokens": 65536,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"extended_thinking",
"code_execution",
- "audio_input",
+ "audio_input"
],
- releaseDate: "2026-02-19",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-01-01",
- resolvedAt: "2026-03-24T11:03:33.071Z",
- baseModelName: null,
+ "releaseDate": "2026-02-19",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-01-01",
+ "resolvedAt": "2026-03-24T11:03:33.071Z",
+ "baseModelName": null
},
"gemini-pro": {
- provider: "google",
- description:
- "Google's first-generation Gemini model for text generation, reasoning, and multi-turn conversation. Superseded by Gemini 1.5 Pro and later models.",
- contextWindow: 32768,
- maxOutputTokens: 8192,
- capabilities: ["tool_use", "streaming", "json_mode"],
- releaseDate: "2023-12-13",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-04-09",
- knowledgeCutoff: "2023-04-01",
- resolvedAt: "2026-03-24T11:03:45.401Z",
- baseModelName: null,
+ "provider": "google",
+ "description": "Google's first-generation Gemini model for text generation, reasoning, and multi-turn conversation. Superseded by Gemini 1.5 Pro and later models.",
+ "contextWindow": 32768,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2023-12-13",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-04-09",
+ "knowledgeCutoff": "2023-04-01",
+ "resolvedAt": "2026-03-24T11:03:45.401Z",
+ "baseModelName": null
},
"gpt-3.5-turbo": {
- provider: "openai",
- description:
- "OpenAI's fast and cost-effective model optimized for chat and instruction-following tasks, now superseded by GPT-4o mini.",
- contextWindow: 16385,
- maxOutputTokens: 4096,
- capabilities: ["tool_use", "streaming", "json_mode", "fine_tunable"],
- releaseDate: "2023-03-01",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2025-09-13",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:03:11.412Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's fast and cost-effective model optimized for chat and instruction-following tasks, now superseded by GPT-4o mini.",
+ "contextWindow": 16385,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "fine_tunable"
+ ],
+ "releaseDate": "2023-03-01",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2025-09-13",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:03:11.412Z",
+ "baseModelName": null
},
"gpt-3.5-turbo-0125": {
- provider: "openai",
- description:
- "A fast and cost-effective GPT-3.5 Turbo snapshot optimized for chat completions, offering improved accuracy for function calling and reduced instances of incomplete responses.",
- contextWindow: 16385,
- maxOutputTokens: 4096,
- capabilities: ["tool_use", "streaming", "json_mode", "fine_tunable"],
- releaseDate: "2024-01-25",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2025-09-13",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:03:11.310Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "A fast and cost-effective GPT-3.5 Turbo snapshot optimized for chat completions, offering improved accuracy for function calling and reduced instances of incomplete responses.",
+ "contextWindow": 16385,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "fine_tunable"
+ ],
+ "releaseDate": "2024-01-25",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2025-09-13",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:03:11.310Z",
+ "baseModelName": null
},
"gpt-3.5-turbo-0301": {
- provider: "openai",
- description:
- "Early snapshot of GPT-3.5 Turbo, OpenAI's first ChatGPT-optimized model for chat completions. Fast and cost-effective for simple tasks but superseded by later revisions.",
- contextWindow: 4096,
- maxOutputTokens: 4096,
- capabilities: ["streaming", "fine_tunable"],
- releaseDate: "2023-03-01",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2024-06-13",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:03:12.060Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "Early snapshot of GPT-3.5 Turbo, OpenAI's first ChatGPT-optimized model for chat completions. Fast and cost-effective for simple tasks but superseded by later revisions.",
+ "contextWindow": 4096,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "streaming",
+ "fine_tunable"
+ ],
+ "releaseDate": "2023-03-01",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2024-06-13",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:03:12.060Z",
+ "baseModelName": null
},
"gpt-3.5-turbo-0613": {
- provider: "openai",
- description:
- "A snapshot of GPT-3.5 Turbo from June 2023, optimized for chat and instruction-following tasks with function calling support.",
- contextWindow: 4096,
- maxOutputTokens: 4096,
- capabilities: ["tool_use", "streaming", "json_mode", "fine_tunable"],
- releaseDate: "2023-06-13",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2024-09-13",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:04:04.463Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "A snapshot of GPT-3.5 Turbo from June 2023, optimized for chat and instruction-following tasks with function calling support.",
+ "contextWindow": 4096,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "fine_tunable"
+ ],
+ "releaseDate": "2023-06-13",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2024-09-13",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:04:04.463Z",
+ "baseModelName": null
},
"gpt-3.5-turbo-1106": {
- provider: "openai",
- description:
- "A dated snapshot of GPT-3.5 Turbo released in November 2023, offering improved instruction following, JSON mode, and parallel function calling over previous GPT-3.5 variants.",
- contextWindow: 16385,
- maxOutputTokens: 4096,
- capabilities: ["tool_use", "streaming", "json_mode", "fine_tunable"],
- releaseDate: "2023-11-06",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2025-09-13",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:04:23.054Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "A dated snapshot of GPT-3.5 Turbo released in November 2023, offering improved instruction following, JSON mode, and parallel function calling over previous GPT-3.5 variants.",
+ "contextWindow": 16385,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "fine_tunable"
+ ],
+ "releaseDate": "2023-11-06",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2025-09-13",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:04:23.054Z",
+ "baseModelName": null
},
"gpt-3.5-turbo-16k": {
- provider: "openai",
- description:
- "Extended context version of GPT-3.5 Turbo with 16K token context window, offering the same capabilities as the base model but able to process longer inputs.",
- contextWindow: 16384,
- maxOutputTokens: 4096,
- capabilities: ["streaming", "json_mode", "fine_tunable", "tool_use"],
- releaseDate: "2023-06-13",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2025-09-13",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:04:36.307Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "Extended context version of GPT-3.5 Turbo with 16K token context window, offering the same capabilities as the base model but able to process longer inputs.",
+ "contextWindow": 16384,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "streaming",
+ "json_mode",
+ "fine_tunable",
+ "tool_use"
+ ],
+ "releaseDate": "2023-06-13",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2025-09-13",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:04:36.307Z",
+ "baseModelName": null
},
"gpt-3.5-turbo-16k-0613": {
- provider: "openai",
- description:
- "Extended context window variant of GPT-3.5 Turbo with 16K token context, snapshot from June 2023. Optimized for chat completions with longer document processing.",
- contextWindow: 16384,
- maxOutputTokens: 4096,
- capabilities: ["streaming", "json_mode", "fine_tunable"],
- releaseDate: "2023-06-13",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2024-09-13",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:04:22.894Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "Extended context window variant of GPT-3.5 Turbo with 16K token context, snapshot from June 2023. Optimized for chat completions with longer document processing.",
+ "contextWindow": 16384,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "streaming",
+ "json_mode",
+ "fine_tunable"
+ ],
+ "releaseDate": "2023-06-13",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2024-09-13",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:04:22.894Z",
+ "baseModelName": null
},
"gpt-3.5-turbo-instruct": {
- provider: "openai",
- description:
- "OpenAI's GPT-3.5 Turbo Instruct is a completions-only model (not chat) optimized for following explicit instructions, replacing the legacy text-davinci-003 model.",
- contextWindow: 4096,
- maxOutputTokens: 4096,
- capabilities: ["fine_tunable"],
- releaseDate: "2023-09-19",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-01-27",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:04:22.309Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's GPT-3.5 Turbo Instruct is a completions-only model (not chat) optimized for following explicit instructions, replacing the legacy text-davinci-003 model.",
+ "contextWindow": 4096,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "fine_tunable"
+ ],
+ "releaseDate": "2023-09-19",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-01-27",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:04:22.309Z",
+ "baseModelName": null
},
"gpt-4": {
- provider: "openai",
- description:
- "OpenAI's flagship large language model that preceded GPT-4o, known for strong reasoning and instruction-following capabilities across a wide range of tasks.",
- contextWindow: 8192,
- maxOutputTokens: 8192,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2023-03-14",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2025-06-06",
- knowledgeCutoff: "2023-12-01",
- resolvedAt: "2026-03-24T11:04:36.773Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's flagship large language model that preceded GPT-4o, known for strong reasoning and instruction-following capabilities across a wide range of tasks.",
+ "contextWindow": 8192,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2023-03-14",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2025-06-06",
+ "knowledgeCutoff": "2023-12-01",
+ "resolvedAt": "2026-03-24T11:04:36.773Z",
+ "baseModelName": null
},
"gpt-4-0125-preview": {
- provider: "openai",
- description:
- "An improved GPT-4 Turbo preview model with better task completion, reduced laziness in code generation, and enhanced instruction following.",
- contextWindow: 128000,
- maxOutputTokens: 4096,
- capabilities: ["tool_use", "streaming", "json_mode"],
- releaseDate: "2024-01-25",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-12-01",
- resolvedAt: "2026-03-24T11:04:54.196Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "An improved GPT-4 Turbo preview model with better task completion, reduced laziness in code generation, and enhanced instruction following.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-01-25",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-12-01",
+ "resolvedAt": "2026-03-24T11:04:54.196Z",
+ "baseModelName": null
},
"gpt-4-0314": {
- provider: "openai",
- description:
- "Original GPT-4 snapshot from March 2023, a large multimodal model (text-only at launch) that was one of OpenAI's first GPT-4 releases. Now deprecated and replaced by newer GPT-4 variants.",
- contextWindow: 8192,
- maxOutputTokens: 4096,
- capabilities: ["streaming"],
- releaseDate: "2023-03-14",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2024-06-13",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:05:14.112Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "Original GPT-4 snapshot from March 2023, a large multimodal model (text-only at launch) that was one of OpenAI's first GPT-4 releases. Now deprecated and replaced by newer GPT-4 variants.",
+ "contextWindow": 8192,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "streaming"
+ ],
+ "releaseDate": "2023-03-14",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2024-06-13",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:05:14.112Z",
+ "baseModelName": null
},
"gpt-4-0613": {
- provider: "openai",
- description:
- "A snapshot of GPT-4 from June 2023, offering strong reasoning and instruction-following capabilities. It was one of the first widely available GPT-4 variants with function calling support.",
- contextWindow: 8192,
- maxOutputTokens: 8192,
- capabilities: ["tool_use", "streaming", "json_mode", "fine_tunable"],
- releaseDate: "2023-06-13",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-06-06",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:05:13.885Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "A snapshot of GPT-4 from June 2023, offering strong reasoning and instruction-following capabilities. It was one of the first widely available GPT-4 variants with function calling support.",
+ "contextWindow": 8192,
+ "maxOutputTokens": 8192,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "fine_tunable"
+ ],
+ "releaseDate": "2023-06-13",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-06-06",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:05:13.885Z",
+ "baseModelName": null
},
"gpt-4-1106-preview": {
- provider: "openai",
- description:
- "GPT-4 Turbo preview model with 128K context window, offering improved instruction following and JSON mode support at reduced cost compared to GPT-4.",
- contextWindow: 128000,
- maxOutputTokens: 4096,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2023-11-06",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-04-01",
- resolvedAt: "2026-03-24T11:05:12.960Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "GPT-4 Turbo preview model with 128K context window, offering improved instruction following and JSON mode support at reduced cost compared to GPT-4.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2023-11-06",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-04-01",
+ "resolvedAt": "2026-03-24T11:05:12.960Z",
+ "baseModelName": null
},
"gpt-4-32k": {
- provider: "openai",
- description:
- "Extended context window variant of GPT-4 with 32,768 token capacity, offering the same capabilities as GPT-4 but able to process longer documents and conversations.",
- contextWindow: 32768,
- maxOutputTokens: 4096,
- capabilities: ["tool_use", "streaming", "json_mode"],
- releaseDate: "2023-03-14",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-06-06",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:05:14.584Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "Extended context window variant of GPT-4 with 32,768 token capacity, offering the same capabilities as GPT-4 but able to process longer documents and conversations.",
+ "contextWindow": 32768,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2023-03-14",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-06-06",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:05:14.584Z",
+ "baseModelName": null
},
"gpt-4-32k-0314": {
- provider: "openai",
- description:
- "Extended context (32k token) variant of the original GPT-4 launch snapshot from March 2024, offering the same capabilities as gpt-4-0314 but with 4x the context window.",
- contextWindow: 32768,
- maxOutputTokens: 4096,
- capabilities: ["streaming"],
- releaseDate: "2023-03-14",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2024-06-13",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:05:32.044Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "Extended context (32k token) variant of the original GPT-4 launch snapshot from March 2024, offering the same capabilities as gpt-4-0314 but with 4x the context window.",
+ "contextWindow": 32768,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "streaming"
+ ],
+ "releaseDate": "2023-03-14",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2024-06-13",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:05:32.044Z",
+ "baseModelName": null
},
"gpt-4-32k-0613": {
- provider: "openai",
- description:
- "Extended context window variant of GPT-4 with 32,768 token context, based on the June 2023 snapshot. Offers the same capabilities as GPT-4 but with 4x the context length.",
- contextWindow: 32768,
- maxOutputTokens: 4096,
- capabilities: ["tool_use", "streaming", "json_mode"],
- releaseDate: "2023-06-13",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-06-06",
- knowledgeCutoff: "2021-09-01",
- resolvedAt: "2026-03-24T11:05:53.070Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "Extended context window variant of GPT-4 with 32,768 token context, based on the June 2023 snapshot. Offers the same capabilities as GPT-4 but with 4x the context length.",
+ "contextWindow": 32768,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2023-06-13",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-06-06",
+ "knowledgeCutoff": "2021-09-01",
+ "resolvedAt": "2026-03-24T11:05:53.070Z",
+ "baseModelName": null
},
"gpt-4-preview": {
- provider: "openai",
- description:
- "GPT-4 Turbo preview model with 128K context window, JSON mode, and parallel function calling. A preview release in the GPT-4 Turbo series, now deprecated in favor of newer models.",
- contextWindow: 128000,
- maxOutputTokens: 4096,
- capabilities: ["tool_use", "streaming", "json_mode"],
- releaseDate: "2023-11-06",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2025-06-06",
- knowledgeCutoff: "2023-04-01",
- resolvedAt: "2026-03-24T11:06:54.248Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "GPT-4 Turbo preview model with 128K context window, JSON mode, and parallel function calling. A preview release in the GPT-4 Turbo series, now deprecated in favor of newer models.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2023-11-06",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2025-06-06",
+ "knowledgeCutoff": "2023-04-01",
+ "resolvedAt": "2026-03-24T11:06:54.248Z",
+ "baseModelName": null
},
"gpt-4-turbo": {
- provider: "openai",
- description:
- "OpenAI's optimized GPT-4 variant offering faster inference and lower cost than the original GPT-4, with vision capabilities and a 128K context window.",
- contextWindow: 128000,
- maxOutputTokens: 4096,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-04-09",
- isHidden: false,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-12-01",
- resolvedAt: "2026-03-24T11:05:51.415Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's optimized GPT-4 variant offering faster inference and lower cost than the original GPT-4, with vision capabilities and a 128K context window.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-04-09",
+ "isHidden": false,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-12-01",
+ "resolvedAt": "2026-03-24T11:05:51.415Z",
+ "baseModelName": null
},
"gpt-4-turbo-2024-04-09": {
- provider: "openai",
- description:
- "OpenAI's optimized GPT-4 variant offering faster inference and lower cost than the original GPT-4, with vision capabilities and a 128K context window.",
- contextWindow: 128000,
- maxOutputTokens: 4096,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-04-09",
- isHidden: false,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-12-01",
- resolvedAt: "2026-03-24T11:05:51.415Z",
- baseModelName: "gpt-4-turbo",
+ "provider": "openai",
+ "description": "OpenAI's optimized GPT-4 variant offering faster inference and lower cost than the original GPT-4, with vision capabilities and a 128K context window.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-04-09",
+ "isHidden": false,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-12-01",
+ "resolvedAt": "2026-03-24T11:05:51.415Z",
+ "baseModelName": "gpt-4-turbo"
},
"gpt-4-turbo-preview": {
- provider: "openai",
- description:
- "An early preview of GPT-4 Turbo with a 128K context window, offering improved instruction following and JSON mode support at reduced cost compared to GPT-4.",
- contextWindow: 128000,
- maxOutputTokens: 4096,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-01-25",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-12-01",
- resolvedAt: "2026-03-24T11:05:52.346Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "An early preview of GPT-4 Turbo with a 128K context window, offering improved instruction following and JSON mode support at reduced cost compared to GPT-4.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-01-25",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-12-01",
+ "resolvedAt": "2026-03-24T11:05:52.346Z",
+ "baseModelName": null
},
"gpt-4-turbo-vision": {
- provider: "openai",
- description:
- "OpenAI's GPT-4 Turbo model with vision capabilities, able to analyze and understand images alongside text. It was a preview model later superseded by GPT-4 Turbo (gpt-4-turbo-2024-04-09) and then GPT-4o.",
- contextWindow: 128000,
- maxOutputTokens: 4096,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2023-11-06",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2024-12-06",
- knowledgeCutoff: "2023-04-01",
- resolvedAt: "2026-03-24T11:06:38.455Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's GPT-4 Turbo model with vision capabilities, able to analyze and understand images alongside text. It was a preview model later superseded by GPT-4 Turbo (gpt-4-turbo-2024-04-09) and then GPT-4o.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2023-11-06",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2024-12-06",
+ "knowledgeCutoff": "2023-04-01",
+ "resolvedAt": "2026-03-24T11:06:38.455Z",
+ "baseModelName": null
},
"gpt-4.1": {
- provider: "openai",
- description:
- "OpenAI's flagship model optimized for coding, instruction following, and tool calling with a 1M token context window. Excels at structured outputs and long-context tasks.",
- contextWindow: 1047576,
- maxOutputTokens: 32768,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-04-14",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:07:00.439Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's flagship model optimized for coding, instruction following, and tool calling with a 1M token context window. Excels at structured outputs and long-context tasks.",
+ "contextWindow": 1047576,
+ "maxOutputTokens": 32768,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-04-14",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:07:00.439Z",
+ "baseModelName": null
},
"gpt-4.1-2025-04-14": {
- provider: "openai",
- description:
- "OpenAI's flagship model optimized for coding, instruction following, and tool calling with a 1M token context window. Excels at structured outputs and long-context tasks.",
- contextWindow: 1047576,
- maxOutputTokens: 32768,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-04-14",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:07:00.439Z",
- baseModelName: "gpt-4.1",
+ "provider": "openai",
+ "description": "OpenAI's flagship model optimized for coding, instruction following, and tool calling with a 1M token context window. Excels at structured outputs and long-context tasks.",
+ "contextWindow": 1047576,
+ "maxOutputTokens": 32768,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-04-14",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:07:00.439Z",
+ "baseModelName": "gpt-4.1"
},
"gpt-4.1-mini": {
- provider: "openai",
- description:
- "A compact, cost-efficient model in OpenAI's GPT-4.1 family that matches or exceeds GPT-4o on many benchmarks while offering nearly half the latency and significantly lower cost.",
- contextWindow: 1000000,
- maxOutputTokens: 32768,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "fine_tunable"],
- releaseDate: "2025-04-14",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:08:14.524Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "A compact, cost-efficient model in OpenAI's GPT-4.1 family that matches or exceeds GPT-4o on many benchmarks while offering nearly half the latency and significantly lower cost.",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 32768,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "fine_tunable"
+ ],
+ "releaseDate": "2025-04-14",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:08:14.524Z",
+ "baseModelName": null
},
"gpt-4.1-mini-2025-04-14": {
- provider: "openai",
- description:
- "A compact, cost-efficient model in OpenAI's GPT-4.1 family that matches or exceeds GPT-4o on many benchmarks while offering nearly half the latency and significantly lower cost.",
- contextWindow: 1000000,
- maxOutputTokens: 32768,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "fine_tunable"],
- releaseDate: "2025-04-14",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:08:14.524Z",
- baseModelName: "gpt-4.1-mini",
+ "provider": "openai",
+ "description": "A compact, cost-efficient model in OpenAI's GPT-4.1 family that matches or exceeds GPT-4o on many benchmarks while offering nearly half the latency and significantly lower cost.",
+ "contextWindow": 1000000,
+ "maxOutputTokens": 32768,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "fine_tunable"
+ ],
+ "releaseDate": "2025-04-14",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:08:14.524Z",
+ "baseModelName": "gpt-4.1-mini"
},
"gpt-4.1-nano": {
- provider: "openai",
- description:
- "OpenAI's fastest and most cost-effective model in the GPT-4.1 family, optimized for low-latency tasks like classification, autocompletion, and lightweight agentic workflows with strong instruction-following and tool-calling capabilities.",
- contextWindow: 1047576,
- maxOutputTokens: 32768,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-04-14",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:08:04.533Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's fastest and most cost-effective model in the GPT-4.1 family, optimized for low-latency tasks like classification, autocompletion, and lightweight agentic workflows with strong instruction-following and tool-calling capabilities.",
+ "contextWindow": 1047576,
+ "maxOutputTokens": 32768,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-04-14",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:08:04.533Z",
+ "baseModelName": null
},
"gpt-4.1-nano-2025-04-14": {
- provider: "openai",
- description:
- "OpenAI's fastest and most cost-effective model in the GPT-4.1 family, optimized for low-latency tasks like classification, autocompletion, and lightweight agentic workflows with strong instruction-following and tool-calling capabilities.",
- contextWindow: 1047576,
- maxOutputTokens: 32768,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-04-14",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:08:04.533Z",
- baseModelName: "gpt-4.1-nano",
+ "provider": "openai",
+ "description": "OpenAI's fastest and most cost-effective model in the GPT-4.1 family, optimized for low-latency tasks like classification, autocompletion, and lightweight agentic workflows with strong instruction-following and tool-calling capabilities.",
+ "contextWindow": 1047576,
+ "maxOutputTokens": 32768,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-04-14",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:08:04.533Z",
+ "baseModelName": "gpt-4.1-nano"
},
"gpt-4.5-preview": {
- provider: "openai",
- description:
- "OpenAI's largest pretrained model before the GPT-5 series, emphasizing broad knowledge, creative writing, and improved emotional intelligence over reasoning-focused models.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-02-27",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2025-07-14",
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:07:57.880Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's largest pretrained model before the GPT-5 series, emphasizing broad knowledge, creative writing, and improved emotional intelligence over reasoning-focused models.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-02-27",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2025-07-14",
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:07:57.880Z",
+ "baseModelName": null
},
"gpt-4.5-preview-2025-02-27": {
- provider: "openai",
- description:
- "OpenAI's largest pretrained model before the GPT-5 series, emphasizing broad knowledge, creative writing, and improved emotional intelligence over reasoning-focused models.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-02-27",
- isHidden: true,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2025-07-14",
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:07:57.880Z",
- baseModelName: "gpt-4.5-preview",
+ "provider": "openai",
+ "description": "OpenAI's largest pretrained model before the GPT-5 series, emphasizing broad knowledge, creative writing, and improved emotional intelligence over reasoning-focused models.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-02-27",
+ "isHidden": true,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2025-07-14",
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:07:57.880Z",
+ "baseModelName": "gpt-4.5-preview"
},
"gpt-4o": {
- provider: "openai",
- description:
- "OpenAI's flagship multimodal model combining strong reasoning with vision, audio, and tool use capabilities at faster speeds and lower cost than GPT-4 Turbo.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: [
+ "provider": "openai",
+ "description": "OpenAI's flagship multimodal model combining strong reasoning with vision, audio, and tool use capabilities at faster speeds and lower cost than GPT-4 Turbo.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"audio_input",
"audio_output",
- "fine_tunable",
+ "fine_tunable"
],
- releaseDate: "2024-05-13",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:07:31.638Z",
- baseModelName: null,
+ "releaseDate": "2024-05-13",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:07:31.638Z",
+ "baseModelName": null
},
"gpt-4o-2024-05-13": {
- provider: "openai",
- description:
- "OpenAI's flagship multimodal model combining strong reasoning with vision, audio, and tool use capabilities at faster speeds and lower cost than GPT-4 Turbo.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: [
+ "provider": "openai",
+ "description": "OpenAI's flagship multimodal model combining strong reasoning with vision, audio, and tool use capabilities at faster speeds and lower cost than GPT-4 Turbo.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"audio_input",
"audio_output",
- "fine_tunable",
+ "fine_tunable"
],
- releaseDate: "2024-05-13",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:07:31.638Z",
- baseModelName: "gpt-4o",
+ "releaseDate": "2024-05-13",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:07:31.638Z",
+ "baseModelName": "gpt-4o"
},
"gpt-4o-2024-08-06": {
- provider: "openai",
- description:
- "OpenAI's flagship multimodal model combining strong reasoning with vision, audio, and tool use capabilities at faster speeds and lower cost than GPT-4 Turbo.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: [
+ "provider": "openai",
+ "description": "OpenAI's flagship multimodal model combining strong reasoning with vision, audio, and tool use capabilities at faster speeds and lower cost than GPT-4 Turbo.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"audio_input",
"audio_output",
- "fine_tunable",
+ "fine_tunable"
],
- releaseDate: "2024-05-13",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:07:31.638Z",
- baseModelName: "gpt-4o",
+ "releaseDate": "2024-05-13",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:07:31.638Z",
+ "baseModelName": "gpt-4o"
},
"gpt-4o-2024-11-20": {
- provider: "openai",
- description:
- "OpenAI's flagship multimodal model combining strong reasoning with vision, audio, and tool use capabilities at faster speeds and lower cost than GPT-4 Turbo.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: [
+ "provider": "openai",
+ "description": "OpenAI's flagship multimodal model combining strong reasoning with vision, audio, and tool use capabilities at faster speeds and lower cost than GPT-4 Turbo.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"audio_input",
"audio_output",
- "fine_tunable",
+ "fine_tunable"
],
- releaseDate: "2024-05-13",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:07:31.638Z",
- baseModelName: "gpt-4o",
+ "releaseDate": "2024-05-13",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:07:31.638Z",
+ "baseModelName": "gpt-4o"
},
"gpt-4o-audio-preview": {
- provider: "openai",
- description:
- "GPT-4o variant with native audio input and output capabilities via the Chat Completions API, supporting both text and audio modalities for conversational and voice-based applications.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: ["audio_input", "audio_output", "tool_use", "streaming"],
- releaseDate: "2024-10-01",
- isHidden: false,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2026-05-07",
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:08:09.590Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "GPT-4o variant with native audio input and output capabilities via the Chat Completions API, supporting both text and audio modalities for conversational and voice-based applications.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
+ "audio_input",
+ "audio_output",
+ "tool_use",
+ "streaming"
+ ],
+ "releaseDate": "2024-10-01",
+ "isHidden": false,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2026-05-07",
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:08:09.590Z",
+ "baseModelName": null
},
"gpt-4o-audio-preview-2024-10-01": {
- provider: "openai",
- description:
- "GPT-4o variant with native audio input and output capabilities via the Chat Completions API, supporting both text and audio modalities for conversational and voice-based applications.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: ["audio_input", "audio_output", "tool_use", "streaming"],
- releaseDate: "2024-10-01",
- isHidden: false,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: "2026-05-07",
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:08:09.590Z",
- baseModelName: "gpt-4o-audio-preview",
+ "provider": "openai",
+ "description": "GPT-4o variant with native audio input and output capabilities via the Chat Completions API, supporting both text and audio modalities for conversational and voice-based applications.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
+ "audio_input",
+ "audio_output",
+ "tool_use",
+ "streaming"
+ ],
+ "releaseDate": "2024-10-01",
+ "isHidden": false,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": "2026-05-07",
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:08:09.590Z",
+ "baseModelName": "gpt-4o-audio-preview"
},
"gpt-4o-mini": {
- provider: "openai",
- description:
- "Fast, affordable small model optimized for focused tasks. Positioned as OpenAI's cost-efficient option with strong performance on benchmarks relative to its size.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "fine_tunable"],
- releaseDate: "2024-07-18",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:09:50.130Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "Fast, affordable small model optimized for focused tasks. Positioned as OpenAI's cost-efficient option with strong performance on benchmarks relative to its size.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "fine_tunable"
+ ],
+ "releaseDate": "2024-07-18",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:09:50.130Z",
+ "baseModelName": null
},
"gpt-4o-mini-2024-07-18": {
- provider: "openai",
- description:
- "Fast, affordable small model optimized for focused tasks. Positioned as OpenAI's cost-efficient option with strong performance on benchmarks relative to its size.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "fine_tunable"],
- releaseDate: "2024-07-18",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:09:50.130Z",
- baseModelName: "gpt-4o-mini",
+ "provider": "openai",
+ "description": "Fast, affordable small model optimized for focused tasks. Positioned as OpenAI's cost-efficient option with strong performance on benchmarks relative to its size.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "fine_tunable"
+ ],
+ "releaseDate": "2024-07-18",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:09:50.130Z",
+ "baseModelName": "gpt-4o-mini"
},
"gpt-4o-realtime-preview": {
- provider: "openai",
- description:
- "OpenAI's real-time multimodal model capable of processing and generating both text and audio over WebRTC or WebSocket, enabling low-latency voice conversations and audio interactions.",
- contextWindow: 128000,
- maxOutputTokens: 4096,
- capabilities: ["audio_input", "audio_output", "tool_use", "streaming"],
- releaseDate: "2024-10-01",
- isHidden: false,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: false,
- deprecationDate: "2026-05-07",
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:09:35.495Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's real-time multimodal model capable of processing and generating both text and audio over WebRTC or WebSocket, enabling low-latency voice conversations and audio interactions.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "audio_input",
+ "audio_output",
+ "tool_use",
+ "streaming"
+ ],
+ "releaseDate": "2024-10-01",
+ "isHidden": false,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2026-05-07",
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:09:35.495Z",
+ "baseModelName": null
},
"gpt-4o-realtime-preview-2024-10-01": {
- provider: "openai",
- description:
- "OpenAI's real-time multimodal model capable of processing and generating both text and audio over WebRTC or WebSocket, enabling low-latency voice conversations and audio interactions.",
- contextWindow: 128000,
- maxOutputTokens: 4096,
- capabilities: ["audio_input", "audio_output", "tool_use", "streaming"],
- releaseDate: "2024-10-01",
- isHidden: false,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: false,
- deprecationDate: "2026-05-07",
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:09:35.495Z",
- baseModelName: "gpt-4o-realtime-preview",
+ "provider": "openai",
+ "description": "OpenAI's real-time multimodal model capable of processing and generating both text and audio over WebRTC or WebSocket, enabling low-latency voice conversations and audio interactions.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 4096,
+ "capabilities": [
+ "audio_input",
+ "audio_output",
+ "tool_use",
+ "streaming"
+ ],
+ "releaseDate": "2024-10-01",
+ "isHidden": false,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2026-05-07",
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:09:35.495Z",
+ "baseModelName": "gpt-4o-realtime-preview"
},
"gpt-5": {
- provider: "openai",
- description:
- "OpenAI's flagship reasoning model released August 2025, featuring a 400K token context window with strong coding, reasoning, and agentic capabilities.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: [
+ "provider": "openai",
+ "description": "OpenAI's flagship reasoning model released August 2025, featuring a 400K token context window with strong coding, reasoning, and agentic capabilities.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"image_generation",
- "code_execution",
+ "code_execution"
],
- releaseDate: "2025-08-07",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-09-30",
- resolvedAt: "2026-03-24T11:09:28.216Z",
- baseModelName: null,
+ "releaseDate": "2025-08-07",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-09-30",
+ "resolvedAt": "2026-03-24T11:09:28.216Z",
+ "baseModelName": null
},
"gpt-5-2025-08-07": {
- provider: "openai",
- description:
- "OpenAI's flagship reasoning model released August 2025, featuring a 400K token context window with strong coding, reasoning, and agentic capabilities.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: [
+ "provider": "openai",
+ "description": "OpenAI's flagship reasoning model released August 2025, featuring a 400K token context window with strong coding, reasoning, and agentic capabilities.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
"vision",
"tool_use",
"streaming",
"json_mode",
"image_generation",
- "code_execution",
+ "code_execution"
],
- releaseDate: "2025-08-07",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-09-30",
- resolvedAt: "2026-03-24T11:09:28.216Z",
- baseModelName: "gpt-5",
+ "releaseDate": "2025-08-07",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-09-30",
+ "resolvedAt": "2026-03-24T11:09:28.216Z",
+ "baseModelName": "gpt-5"
},
"gpt-5-chat-latest": {
- provider: "openai",
- description:
- "Non-reasoning GPT-5 model used in ChatGPT, optimized for conversational tasks. Supports text and image inputs with function calling and structured outputs.",
- contextWindow: 128000,
- maxOutputTokens: 16384,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-08-07",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-09-30",
- resolvedAt: "2026-03-24T11:09:24.834Z",
- baseModelName: "gpt-5-chat",
+ "provider": "openai",
+ "description": "Non-reasoning GPT-5 model used in ChatGPT, optimized for conversational tasks. Supports text and image inputs with function calling and structured outputs.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 16384,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-08-07",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-09-30",
+ "resolvedAt": "2026-03-24T11:09:24.834Z",
+ "baseModelName": "gpt-5-chat"
},
"gpt-5-mini": {
- provider: "openai",
- description:
- "A faster, more cost-efficient version of GPT-5 designed for well-defined tasks and precise prompts. Supports reasoning with configurable effort levels and offers reduced latency compared to the full GPT-5 model.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-08-07",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-05-31",
- resolvedAt: "2026-03-24T11:09:42.822Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "A faster, more cost-efficient version of GPT-5 designed for well-defined tasks and precise prompts. Supports reasoning with configurable effort levels and offers reduced latency compared to the full GPT-5 model.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-08-07",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-05-31",
+ "resolvedAt": "2026-03-24T11:09:42.822Z",
+ "baseModelName": null
},
"gpt-5-mini-2025-08-07": {
- provider: "openai",
- description:
- "A faster, more cost-efficient version of GPT-5 designed for well-defined tasks and precise prompts. Supports reasoning with configurable effort levels and offers reduced latency compared to the full GPT-5 model.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-08-07",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-05-31",
- resolvedAt: "2026-03-24T11:09:42.822Z",
- baseModelName: "gpt-5-mini",
+ "provider": "openai",
+ "description": "A faster, more cost-efficient version of GPT-5 designed for well-defined tasks and precise prompts. Supports reasoning with configurable effort levels and offers reduced latency compared to the full GPT-5 model.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-08-07",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-05-31",
+ "resolvedAt": "2026-03-24T11:09:42.822Z",
+ "baseModelName": "gpt-5-mini"
},
"gpt-5-nano": {
- provider: "openai",
- description:
- "The smallest and fastest variant in the GPT-5 family, optimized for developer tools, rapid interactions, and ultra-low latency environments. Best suited for classification, data extraction, ranking, and sub-agent tasks.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-08-07",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-05-31",
- resolvedAt: "2026-03-24T11:11:24.884Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "The smallest and fastest variant in the GPT-5 family, optimized for developer tools, rapid interactions, and ultra-low latency environments. Best suited for classification, data extraction, ranking, and sub-agent tasks.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-08-07",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-05-31",
+ "resolvedAt": "2026-03-24T11:11:24.884Z",
+ "baseModelName": null
},
"gpt-5-nano-2025-08-07": {
- provider: "openai",
- description:
- "The smallest and fastest variant in the GPT-5 family, optimized for developer tools, rapid interactions, and ultra-low latency environments. Best suited for classification, data extraction, ranking, and sub-agent tasks.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-08-07",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-05-31",
- resolvedAt: "2026-03-24T11:11:24.884Z",
- baseModelName: "gpt-5-nano",
+ "provider": "openai",
+ "description": "The smallest and fastest variant in the GPT-5 family, optimized for developer tools, rapid interactions, and ultra-low latency environments. Best suited for classification, data extraction, ranking, and sub-agent tasks.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-08-07",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-05-31",
+ "resolvedAt": "2026-03-24T11:11:24.884Z",
+ "baseModelName": "gpt-5-nano"
},
"gpt-5-pro": {
- provider: "openai",
- description:
- "OpenAI's enhanced GPT-5 variant optimized for complex tasks requiring step-by-step reasoning, with reduced hallucination and improved code quality compared to the base GPT-5.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-10-06",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-10-01",
- resolvedAt: "2026-03-24T11:11:37.048Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's enhanced GPT-5 variant optimized for complex tasks requiring step-by-step reasoning, with reduced hallucination and improved code quality compared to the base GPT-5.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-10-06",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-10-01",
+ "resolvedAt": "2026-03-24T11:11:37.048Z",
+ "baseModelName": null
},
"gpt-5-pro-2025-10-06": {
- provider: "openai",
- description:
- "OpenAI's enhanced GPT-5 variant optimized for complex tasks requiring step-by-step reasoning, with reduced hallucination and improved code quality compared to the base GPT-5.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-10-06",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-10-01",
- resolvedAt: "2026-03-24T11:11:37.048Z",
- baseModelName: "gpt-5-pro",
+ "provider": "openai",
+ "description": "OpenAI's enhanced GPT-5 variant optimized for complex tasks requiring step-by-step reasoning, with reduced hallucination and improved code quality compared to the base GPT-5.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-10-06",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-10-01",
+ "resolvedAt": "2026-03-24T11:11:37.048Z",
+ "baseModelName": "gpt-5-pro"
},
"gpt-5.1": {
- provider: "openai",
- description:
- "GPT-5.1 is OpenAI's frontier-grade model in the GPT-5 series, offering adaptive reasoning with configurable effort levels, improved coding and math performance, and a more natural conversational style compared to GPT-5.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-11-13",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-09-30",
- resolvedAt: "2026-03-24T11:11:47.327Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "GPT-5.1 is OpenAI's frontier-grade model in the GPT-5 series, offering adaptive reasoning with configurable effort levels, improved coding and math performance, and a more natural conversational style compared to GPT-5.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-11-13",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-09-30",
+ "resolvedAt": "2026-03-24T11:11:47.327Z",
+ "baseModelName": null
},
"gpt-5.1-2025-11-13": {
- provider: "openai",
- description:
- "GPT-5.1 is OpenAI's frontier-grade model in the GPT-5 series, offering adaptive reasoning with configurable effort levels, improved coding and math performance, and a more natural conversational style compared to GPT-5.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2025-11-13",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-09-30",
- resolvedAt: "2026-03-24T11:11:47.327Z",
- baseModelName: "gpt-5.1",
+ "provider": "openai",
+ "description": "GPT-5.1 is OpenAI's frontier-grade model in the GPT-5 series, offering adaptive reasoning with configurable effort levels, improved coding and math performance, and a more natural conversational style compared to GPT-5.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2025-11-13",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-09-30",
+ "resolvedAt": "2026-03-24T11:11:47.327Z",
+ "baseModelName": "gpt-5.1"
},
"gpt-5.2": {
- provider: "openai",
- description:
- "OpenAI's flagship multimodal model released December 2025, excelling at long-context reasoning, agentic tool use, software engineering, and professional knowledge work. Available in Instant, Thinking, and Pro variants.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-12-11",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:11:13.129Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's flagship multimodal model released December 2025, excelling at long-context reasoning, agentic tool use, software engineering, and professional knowledge work. Available in Instant, Thinking, and Pro variants.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-12-11",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:11:13.129Z",
+ "baseModelName": null
},
"gpt-5.2-2025-12-11": {
- provider: "openai",
- description:
- "OpenAI's flagship multimodal model released December 2025, excelling at long-context reasoning, agentic tool use, software engineering, and professional knowledge work. Available in Instant, Thinking, and Pro variants.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-12-11",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:11:13.129Z",
- baseModelName: "gpt-5.2",
+ "provider": "openai",
+ "description": "OpenAI's flagship multimodal model released December 2025, excelling at long-context reasoning, agentic tool use, software engineering, and professional knowledge work. Available in Instant, Thinking, and Pro variants.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-12-11",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:11:13.129Z",
+ "baseModelName": "gpt-5.2"
},
"gpt-5.2-pro": {
- provider: "openai",
- description:
- "OpenAI's previous pro-tier reasoning model optimized for complex professional work requiring step-by-step reasoning, instruction following, and accuracy in high-stakes use cases. Superseded by GPT-5.4 pro.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "extended_thinking"],
- releaseDate: "2025-12-11",
- isHidden: false,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:11:12.711Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's previous pro-tier reasoning model optimized for complex professional work requiring step-by-step reasoning, instruction following, and accuracy in high-stakes use cases. Superseded by GPT-5.4 pro.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-12-11",
+ "isHidden": false,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:11:12.711Z",
+ "baseModelName": null
},
"gpt-5.2-pro-2025-12-11": {
- provider: "openai",
- description:
- "OpenAI's previous pro-tier reasoning model optimized for complex professional work requiring step-by-step reasoning, instruction following, and accuracy in high-stakes use cases. Superseded by GPT-5.4 pro.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "extended_thinking"],
- releaseDate: "2025-12-11",
- isHidden: false,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:11:12.711Z",
- baseModelName: "gpt-5.2-pro",
+ "provider": "openai",
+ "description": "OpenAI's previous pro-tier reasoning model optimized for complex professional work requiring step-by-step reasoning, instruction following, and accuracy in high-stakes use cases. Superseded by GPT-5.4 pro.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-12-11",
+ "isHidden": false,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:11:12.711Z",
+ "baseModelName": "gpt-5.2-pro"
},
"gpt-5.4": {
- provider: "openai",
- description:
- "OpenAI's most capable frontier model as of March 2026, featuring state-of-the-art coding, native computer-use capabilities, and a 1M-token context window for professional and agentic workflows.",
- contextWindow: 1050000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "code_execution"],
- releaseDate: "2026-03-05",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:12:09.220Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's most capable frontier model as of March 2026, featuring state-of-the-art coding, native computer-use capabilities, and a 1M-token context window for professional and agentic workflows.",
+ "contextWindow": 1050000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "code_execution"
+ ],
+ "releaseDate": "2026-03-05",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:12:09.220Z",
+ "baseModelName": null
},
"gpt-5.4-2026-03-05": {
- provider: "openai",
- description:
- "OpenAI's most capable frontier model as of March 2026, featuring state-of-the-art coding, native computer-use capabilities, and a 1M-token context window for professional and agentic workflows.",
- contextWindow: 1050000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "code_execution"],
- releaseDate: "2026-03-05",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:12:09.220Z",
- baseModelName: "gpt-5.4",
+ "provider": "openai",
+ "description": "OpenAI's most capable frontier model as of March 2026, featuring state-of-the-art coding, native computer-use capabilities, and a 1M-token context window for professional and agentic workflows.",
+ "contextWindow": 1050000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "code_execution"
+ ],
+ "releaseDate": "2026-03-05",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:12:09.220Z",
+ "baseModelName": "gpt-5.4"
},
"gpt-5.4-mini": {
- provider: "openai",
- description:
- "OpenAI's fast and efficient small model from the GPT-5.4 family, designed for high-volume workloads. Approaches GPT-5.4 performance on coding and reasoning while running over 2x faster than GPT-5 mini.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2026-03-17",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:12:35.473Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's fast and efficient small model from the GPT-5.4 family, designed for high-volume workloads. Approaches GPT-5.4 performance on coding and reasoning while running over 2x faster than GPT-5 mini.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2026-03-17",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:12:35.473Z",
+ "baseModelName": null
},
"gpt-5.4-mini-2026-03-17": {
- provider: "openai",
- description:
- "OpenAI's fast and efficient small model from the GPT-5.4 family, designed for high-volume workloads. Approaches GPT-5.4 performance on coding and reasoning while running over 2x faster than GPT-5 mini.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2026-03-17",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:12:35.473Z",
- baseModelName: "gpt-5.4-mini",
+ "provider": "openai",
+ "description": "OpenAI's fast and efficient small model from the GPT-5.4 family, designed for high-volume workloads. Approaches GPT-5.4 performance on coding and reasoning while running over 2x faster than GPT-5 mini.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2026-03-17",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:12:35.473Z",
+ "baseModelName": "gpt-5.4-mini"
},
"gpt-5.4-nano": {
- provider: "openai",
- description:
- "OpenAI's cheapest GPT-5.4-class model optimized for simple high-volume tasks like classification, data extraction, ranking, and sub-agent delegation in agentic workflows.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2026-03-17",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:12:52.285Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's cheapest GPT-5.4-class model optimized for simple high-volume tasks like classification, data extraction, ranking, and sub-agent delegation in agentic workflows.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2026-03-17",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:12:52.285Z",
+ "baseModelName": null
},
"gpt-5.4-nano-2026-03-17": {
- provider: "openai",
- description:
- "OpenAI's cheapest GPT-5.4-class model optimized for simple high-volume tasks like classification, data extraction, ranking, and sub-agent delegation in agentic workflows.",
- contextWindow: 400000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2026-03-17",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:12:52.285Z",
- baseModelName: "gpt-5.4-nano",
+ "provider": "openai",
+ "description": "OpenAI's cheapest GPT-5.4-class model optimized for simple high-volume tasks like classification, data extraction, ranking, and sub-agent delegation in agentic workflows.",
+ "contextWindow": 400000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2026-03-17",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:12:52.285Z",
+ "baseModelName": "gpt-5.4-nano"
},
"gpt-5.4-pro": {
- provider: "openai",
- description:
- "OpenAI's highest-capability GPT-5.4 variant, using additional compute for harder problems. Available via Responses API only, designed for complex reasoning, coding, and agentic workflows.",
- contextWindow: 1050000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "extended_thinking"],
- releaseDate: "2026-03-05",
- isHidden: false,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:12:56.903Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's highest-capability GPT-5.4 variant, using additional compute for harder problems. Available via Responses API only, designed for complex reasoning, coding, and agentic workflows.",
+ "contextWindow": 1050000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "extended_thinking"
+ ],
+ "releaseDate": "2026-03-05",
+ "isHidden": false,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:12:56.903Z",
+ "baseModelName": null
},
"gpt-5.4-pro-2026-03-05": {
- provider: "openai",
- description:
- "OpenAI's highest-capability GPT-5.4 variant, using additional compute for harder problems. Available via Responses API only, designed for complex reasoning, coding, and agentic workflows.",
- contextWindow: 1050000,
- maxOutputTokens: 128000,
- capabilities: ["vision", "tool_use", "streaming", "extended_thinking"],
- releaseDate: "2026-03-05",
- isHidden: false,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: "2025-08-31",
- resolvedAt: "2026-03-24T11:12:56.903Z",
- baseModelName: "gpt-5.4-pro",
- },
- o1: {
- provider: "openai",
- description:
- "OpenAI's reasoning model designed for complex tasks requiring multi-step logical thinking, excelling at math, science, and coding problems.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-12-17",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:12:23.948Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's highest-capability GPT-5.4 variant, using additional compute for harder problems. Available via Responses API only, designed for complex reasoning, coding, and agentic workflows.",
+ "contextWindow": 1050000,
+ "maxOutputTokens": 128000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "extended_thinking"
+ ],
+ "releaseDate": "2026-03-05",
+ "isHidden": false,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-08-31",
+ "resolvedAt": "2026-03-24T11:12:56.903Z",
+ "baseModelName": "gpt-5.4-pro"
+ },
+ "o1": {
+ "provider": "openai",
+ "description": "OpenAI's reasoning model designed for complex tasks requiring multi-step logical thinking, excelling at math, science, and coding problems.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-12-17",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:12:23.948Z",
+ "baseModelName": null
},
"o1-2024-12-17": {
- provider: "openai",
- description:
- "OpenAI's reasoning model designed for complex tasks requiring multi-step logical thinking, excelling at math, science, and coding problems.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode"],
- releaseDate: "2024-12-17",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:12:23.948Z",
- baseModelName: "o1",
+ "provider": "openai",
+ "description": "OpenAI's reasoning model designed for complex tasks requiring multi-step logical thinking, excelling at math, science, and coding problems.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-12-17",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:12:23.948Z",
+ "baseModelName": "o1"
},
"o1-mini": {
- provider: "openai",
- description:
- "A smaller, faster, and cheaper reasoning model in OpenAI's o1 series, optimized for coding, math, and science tasks requiring multi-step reasoning.",
- contextWindow: 128000,
- maxOutputTokens: 65536,
- capabilities: ["streaming", "json_mode"],
- releaseDate: "2024-09-12",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-06-30",
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:12:37.030Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "A smaller, faster, and cheaper reasoning model in OpenAI's o1 series, optimized for coding, math, and science tasks requiring multi-step reasoning.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 65536,
+ "capabilities": [
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-09-12",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-06-30",
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:12:37.030Z",
+ "baseModelName": null
},
"o1-mini-2024-09-12": {
- provider: "openai",
- description:
- "A smaller, faster, and cheaper reasoning model in OpenAI's o1 series, optimized for coding, math, and science tasks requiring multi-step reasoning.",
- contextWindow: 128000,
- maxOutputTokens: 65536,
- capabilities: ["streaming", "json_mode"],
- releaseDate: "2024-09-12",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-06-30",
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:12:37.030Z",
- baseModelName: "o1-mini",
+ "provider": "openai",
+ "description": "A smaller, faster, and cheaper reasoning model in OpenAI's o1 series, optimized for coding, math, and science tasks requiring multi-step reasoning.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 65536,
+ "capabilities": [
+ "streaming",
+ "json_mode"
+ ],
+ "releaseDate": "2024-09-12",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-06-30",
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:12:37.030Z",
+ "baseModelName": "o1-mini"
},
"o1-preview": {
- provider: "openai",
- description:
- "OpenAI's first reasoning model using chain-of-thought to solve complex problems in science, coding, and math. Predecessor to o1 and o3 series.",
- contextWindow: 128000,
- maxOutputTokens: 32768,
- capabilities: ["streaming"],
- releaseDate: "2024-09-12",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-10-31",
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:12:59.198Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's first reasoning model using chain-of-thought to solve complex problems in science, coding, and math. Predecessor to o1 and o3 series.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 32768,
+ "capabilities": [
+ "streaming"
+ ],
+ "releaseDate": "2024-09-12",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-10-31",
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:12:59.198Z",
+ "baseModelName": null
},
"o1-preview-2024-09-12": {
- provider: "openai",
- description:
- "OpenAI's first reasoning model using chain-of-thought to solve complex problems in science, coding, and math. Predecessor to o1 and o3 series.",
- contextWindow: 128000,
- maxOutputTokens: 32768,
- capabilities: ["streaming"],
- releaseDate: "2024-09-12",
- isHidden: true,
- supportsStructuredOutput: false,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: "2025-10-31",
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:12:59.198Z",
- baseModelName: "o1-preview",
+ "provider": "openai",
+ "description": "OpenAI's first reasoning model using chain-of-thought to solve complex problems in science, coding, and math. Predecessor to o1 and o3 series.",
+ "contextWindow": 128000,
+ "maxOutputTokens": 32768,
+ "capabilities": [
+ "streaming"
+ ],
+ "releaseDate": "2024-09-12",
+ "isHidden": true,
+ "supportsStructuredOutput": false,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": "2025-10-31",
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:12:59.198Z",
+ "baseModelName": "o1-preview"
},
"o1-pro": {
- provider: "openai",
- description:
- "A version of OpenAI's o1 reasoning model that uses significantly more compute to deliver better, more consistent answers on complex reasoning tasks in science, coding, and math.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["vision", "tool_use", "json_mode", "extended_thinking"],
- releaseDate: "2025-03-19",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:13:57.532Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "A version of OpenAI's o1 reasoning model that uses significantly more compute to deliver better, more consistent answers on complex reasoning tasks in science, coding, and math.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-03-19",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:13:57.532Z",
+ "baseModelName": null
},
"o1-pro-2025-03-19": {
- provider: "openai",
- description:
- "A version of OpenAI's o1 reasoning model that uses significantly more compute to deliver better, more consistent answers on complex reasoning tasks in science, coding, and math.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["vision", "tool_use", "json_mode", "extended_thinking"],
- releaseDate: "2025-03-19",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: "2023-10-01",
- resolvedAt: "2026-03-24T11:13:57.532Z",
- baseModelName: "o1-pro",
- },
- o3: {
- provider: "openai",
- description:
- "OpenAI's advanced reasoning model designed for complex tasks requiring deep reasoning, excelling at software engineering, mathematics, scientific reasoning, and visual reasoning tasks.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-04-16",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:14:04.906Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "A version of OpenAI's o1 reasoning model that uses significantly more compute to deliver better, more consistent answers on complex reasoning tasks in science, coding, and math.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-03-19",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2023-10-01",
+ "resolvedAt": "2026-03-24T11:13:57.532Z",
+ "baseModelName": "o1-pro"
+ },
+ "o3": {
+ "provider": "openai",
+ "description": "OpenAI's advanced reasoning model designed for complex tasks requiring deep reasoning, excelling at software engineering, mathematics, scientific reasoning, and visual reasoning tasks.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-04-16",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:14:04.906Z",
+ "baseModelName": null
},
"o3-2025-04-16": {
- provider: "openai",
- description:
- "OpenAI's advanced reasoning model designed for complex tasks requiring deep reasoning, excelling at software engineering, mathematics, scientific reasoning, and visual reasoning tasks.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-04-16",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:14:04.906Z",
- baseModelName: "o3",
+ "provider": "openai",
+ "description": "OpenAI's advanced reasoning model designed for complex tasks requiring deep reasoning, excelling at software engineering, mathematics, scientific reasoning, and visual reasoning tasks.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-04-16",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:14:04.906Z",
+ "baseModelName": "o3"
},
"o3-mini": {
- provider: "openai",
- description:
- "OpenAI's compact reasoning model optimized for STEM tasks, offering strong performance in math, science, and coding at lower cost than o3.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-01-31",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-01-01",
- resolvedAt: "2026-03-24T11:13:33.788Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's compact reasoning model optimized for STEM tasks, offering strong performance in math, science, and coding at lower cost than o3.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-01-31",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-01-01",
+ "resolvedAt": "2026-03-24T11:13:33.788Z",
+ "baseModelName": null
},
"o3-mini-2025-01-31": {
- provider: "openai",
- description:
- "OpenAI's compact reasoning model optimized for STEM tasks, offering strong performance in math, science, and coding at lower cost than o3.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-01-31",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2025-01-01",
- resolvedAt: "2026-03-24T11:13:33.788Z",
- baseModelName: "o3-mini",
+ "provider": "openai",
+ "description": "OpenAI's compact reasoning model optimized for STEM tasks, offering strong performance in math, science, and coding at lower cost than o3.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-01-31",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2025-01-01",
+ "resolvedAt": "2026-03-24T11:13:33.788Z",
+ "baseModelName": "o3-mini"
},
"o3-pro": {
- provider: "openai",
- description:
- "OpenAI's most reliable reasoning model, a version of o3 designed to think longer and provide more consistently accurate answers for challenging math, science, and coding problems.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-06-10",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:14:10.900Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's most reliable reasoning model, a version of o3 designed to think longer and provide more consistently accurate answers for challenging math, science, and coding problems.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-06-10",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:14:10.900Z",
+ "baseModelName": null
},
"o3-pro-2025-06-10": {
- provider: "openai",
- description:
- "OpenAI's most reliable reasoning model, a version of o3 designed to think longer and provide more consistently accurate answers for challenging math, science, and coding problems.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-06-10",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: false,
- supportsStreamingToolCalls: false,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:14:10.900Z",
- baseModelName: "o3-pro",
+ "provider": "openai",
+ "description": "OpenAI's most reliable reasoning model, a version of o3 designed to think longer and provide more consistently accurate answers for challenging math, science, and coding problems.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-06-10",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": false,
+ "supportsStreamingToolCalls": false,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:14:10.900Z",
+ "baseModelName": "o3-pro"
},
"o4-mini": {
- provider: "openai",
- description:
- "OpenAI's small reasoning model optimized for fast, cost-efficient reasoning with strong performance in math, coding, and visual tasks.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-04-16",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:14:16.050Z",
- baseModelName: null,
+ "provider": "openai",
+ "description": "OpenAI's small reasoning model optimized for fast, cost-efficient reasoning with strong performance in math, coding, and visual tasks.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-04-16",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:14:16.050Z",
+ "baseModelName": null
},
"o4-mini-2025-04-16": {
- provider: "openai",
- description:
- "OpenAI's small reasoning model optimized for fast, cost-efficient reasoning with strong performance in math, coding, and visual tasks.",
- contextWindow: 200000,
- maxOutputTokens: 100000,
- capabilities: ["vision", "tool_use", "streaming", "json_mode", "extended_thinking"],
- releaseDate: "2025-04-16",
- isHidden: false,
- supportsStructuredOutput: true,
- supportsParallelToolCalls: true,
- supportsStreamingToolCalls: true,
- deprecationDate: null,
- knowledgeCutoff: "2024-06-01",
- resolvedAt: "2026-03-24T11:14:16.050Z",
- baseModelName: "o4-mini",
- },
+ "provider": "openai",
+ "description": "OpenAI's small reasoning model optimized for fast, cost-efficient reasoning with strong performance in math, coding, and visual tasks.",
+ "contextWindow": 200000,
+ "maxOutputTokens": 100000,
+ "capabilities": [
+ "vision",
+ "tool_use",
+ "streaming",
+ "json_mode",
+ "extended_thinking"
+ ],
+ "releaseDate": "2025-04-16",
+ "isHidden": false,
+ "supportsStructuredOutput": true,
+ "supportsParallelToolCalls": true,
+ "supportsStreamingToolCalls": true,
+ "deprecationDate": null,
+ "knowledgeCutoff": "2024-06-01",
+ "resolvedAt": "2026-03-24T11:14:16.050Z",
+ "baseModelName": "o4-mini"
+ }
};
-
+