Skip to content

Pull requests: NVIDIA/TensorRT-LLM

New pull request New
946 Open 11,820 Closed
946 Open 11,820 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[None][perf] Add BSX multi-tier CuTe DSL top-k decode kernels (stacked on #16457)
#16877 opened Jul 26, 2026 by longcheng-nv Collaborator • Draft
5 of 7 tasks
@longcheng-nv
[None][perf] optimize native V2 KV event production
#16876 opened Jul 26, 2026 by alec-flowers Collaborator • Draft
@alec-flowers
[None][perf] Address inter-iter idle times
#16875 opened Jul 26, 2026 by brb-nv Collaborator Loading…
1 task done
@brb-nv
[None][fix] Do not synthesize mrope_config for text-only prompts
#16874 opened Jul 26, 2026 by lingjiew • Draft
@lingjiew
[None][perf] serve: always use msgspec msgpack for disagg orchestrator->worker body
#16873 opened Jul 25, 2026 by Tabrizian Member Loading…
1 task done
@Tabrizian
8
[https://nvbugs/6480621][feat] add disaggregated lifecycle diagnostics
#16872 opened Jul 25, 2026 by chienchunhung Collaborator • Draft
@chienchunhung
14
[https://nvbugs/6506918][fix] Reject the incompatible configuration up front in Linear.__init__ with an…
#16870 opened Jul 25, 2026 by trtllm-agent Collaborator Loading…
2 tasks done
@QiJune @trtllm-agent
3
[None][feat] publish native V2 KV cache events
#16869 opened Jul 25, 2026 by alec-flowers Collaborator • Draft
@alec-flowers
[https://nvbugs/6507114][fix] Extend the per-model qwen3.5_moe_35b.yaml with…
#16868 opened Jul 25, 2026 by trtllm-agent Collaborator Loading…
2 tasks done
@suyoggupta @trtllm-agent
1
[https://nvbugs/6240584][fix] Qwen3ToolParser: bare-JSON fallback for reasoning-preceded tool calls
#16866 opened Jul 25, 2026 by JunyiXu-nv Collaborator Loading…
1 task done
@JunyiXu-nv
8
[TRTLLM-14609][chore] Remove legacy MoE path in DenseGEMMFusedMoE
#16865 opened Jul 25, 2026 by xxi-nv Collaborator • Draft
3 tasks done
@xxi-nv
[TRTLLM-14609][chore] Remove legacy MoE path in DeepGemmFusedMoE
#16864 opened Jul 25, 2026 by xxi-nv Collaborator • Draft
3 tasks done
@xxi-nv
[TRTLLM-14609][chore] Remove legacy MoE path in CuteDslFusedMoE
#16863 opened Jul 25, 2026 by xxi-nv Collaborator • Draft
3 tasks done
@xxi-nv
[TRTLLM-14609][chore] Remove legacy MoE path in TRTLLMGenFusedMoE
#16862 opened Jul 25, 2026 by xxi-nv Collaborator • Draft
3 tasks done
@xxi-nv
[TRTLLM-14609][chore] Remove legacy MoE path in CutlassFusedMoE
#16861 opened Jul 25, 2026 by xxi-nv Collaborator • Draft
3 tasks done
@xxi-nv
[None][perf] Fuse MiniMax-M3 MoE routing
#16859 opened Jul 25, 2026 by peihu-nv Collaborator Loading…
1 task done
@peihu-nv
3
[https://nvbugs/6506920][fix] Add stderr, stderr_margin_sigmas fields to HypothesisTestingParams…
#16858 opened Jul 25, 2026 by trtllm-agent Collaborator Loading…
2 tasks done
@jiaganc @trtllm-agent
3
[None][perf] Avoid paged MSA K/V materialization during prefill
#16857 opened Jul 25, 2026 by peihu-nv Collaborator • Draft
1 task done
@peihu-nv
3
[None][perf] Avoid Index-K cache materialization for MSA
#16856 opened Jul 25, 2026 by peihu-nv Collaborator Loading…
1 task done
@peihu-nv
4
[None][test] Stabilize scaffolding OpenAI worker tests
#16853 opened Jul 24, 2026 by Mgluhovskoi Contributor • Draft
@Mgluhovskoi
3
[None][perf] Fuse MiniMax-M3 prefill projections api-compatible Accepted LLM API contract change that is backwards-compatible
#16852 opened Jul 24, 2026 by peihu-nv Collaborator • Draft
2 tasks done
@peihu-nv
[None][fix] Increase max top logprobs limit
#16851 opened Jul 24, 2026 by yibinl-nvidia Collaborator Loading…
1 task
@yibinl-nvidia
4
[https://nvbugs/6506990][fix] Don't treat NVLE-only as CC enabled
#16850 opened Jul 24, 2026 by dhansen-nvidia Collaborator Loading…
1 task done
@dhansen-nvidia
5
[None][perf] fp8 block scale quant fusion in SM90 Cutlass MoE
#16849 opened Jul 24, 2026 by amukkara Collaborator Loading…
1 task done
@amukkara
5
[None][fix] Update VisualGen test CODEOWNERS
#16848 opened Jul 24, 2026 by yibinl-nvidia Collaborator Loading…
1 task done
@yibinl-nvidia
4
ProTip! Filter pull requests by the default branch with base:main.