Pull requests: NVIDIA/TensorRT-LLM
Pull requests list
[https://nvbugs/6071380][fix] Update the invalid dynamo urls in doc.
Cherry-pick
It's a label that applies to Cherry-pick PR.
#13038
opened Apr 14, 2026 by
nv-guomingz
Collaborator
Loading…
[https://nvbugs/6071380][fix] Update the invalid dynamo urls in doc.
#13037
opened Apr 14, 2026 by
nv-guomingz
Collaborator
Loading…
1 task done
[TRTLLM-11872][perf] Optimize async media loading and video frame decoding in trtllm-serve
#13034
opened Apr 14, 2026 by
yechank-nvidia
Collaborator
Loading…
[None][feat] Update rms_norm + fp4_qaunt kernel supporting more dim
#13033
opened Apr 14, 2026 by
Wanli-Jiang
Collaborator
•
Draft
1 task done
[None][feat] Optimize nemotron-h from python level
#13032
opened Apr 14, 2026 by
Wanli-Jiang
Collaborator
Loading…
1 task done
[None][chore] Add aggregated benchmark in slurm.
#13030
opened Apr 14, 2026 by
dominicshanshan
Collaborator
Loading…
1 task done
[None][refactor] Batch addSequence with two-phase claim and unified VSWA/non-reuse support
#13029
opened Apr 14, 2026 by
liji-nv
Collaborator
Loading…
1 task done
[None][test] Add sync_qa_tests Jenkins script and sync-qa-tests skill
#13028
opened Apr 14, 2026 by
xinhe-nv
Collaborator
Loading…
1 task done
[None][fix] handle broken symlinks in build_wheel.py install_file
#13027
opened Apr 14, 2026 by
zhenhuaw-me
Member
Loading…
1 task done
[None][fix] fix Wan unit tests
VisualGen
#13026
opened Apr 14, 2026 by
zhenhuaw-me
Member
Loading…
1 task done
[None][fix] Fix chunked prefill API contract for nemotron nano VL
#13025
opened Apr 14, 2026 by
2ez4bz
Collaborator
Loading…
1 task done
[#13023][infra] Add Docker BuildKit registry cache to speed up image …
Community want to contribute
PRs initiated from Community
#13024
opened Apr 14, 2026 by
sandeshwani
Loading…
7 tasks
[#13021][infra] Add concurrency control and permissions to PR check workflow
Community want to contribute
PRs initiated from Community
#13022
opened Apr 14, 2026 by
sandeshwani
Loading…
1 of 3 tasks
[Draft][USE CI TO TEST] add back kernel selection logic for spec-dec tree
#13019
opened Apr 14, 2026 by
pengbowang-nv
Collaborator
•
Draft
1 task
[https://nvbugs/6050483][fix] Pin diffusers to 0.37.1 to fix UniPC scheduler device mismatch
#13017
opened Apr 14, 2026 by
chang-l
Collaborator
Loading…
4 tasks done
[feat][None] Fused moe all-reduce routed scaling factor + quant support
Community want to contribute
PRs initiated from Community
#13015
opened Apr 13, 2026 by
murphymatt
Loading…
1 task done
[None][perf] Gemma4 MoE: fused decode kernels, router fence, multi-stream + CUDA graph fixes
#13013
opened Apr 13, 2026 by
nvchenghaoz
Collaborator
•
Draft
4 tasks
[None][perf] AutoDeploy: reduce C++ dispatch overhead in decode scheduling loop
#13012
opened Apr 13, 2026 by
nvchenghaoz
Collaborator
Loading…
2 tasks done
[None][perf] triton paged attention: non-pow2 head_dim, decode speedup, logit cap
#13010
opened Apr 13, 2026 by
nvchenghaoz
Collaborator
Loading…
2 tasks done
[https://nvbugs/6070878][fix] Skip gemma3 fp8 test only on L40S
#13009
opened Apr 13, 2026 by
brb-nv
Collaborator
Loading…
1 task done
[None][infra] Support nv sa benchmark in CI Perf Test
#13004
opened Apr 13, 2026 by
chenfeiz0326
Collaborator
Loading…
1 task done
ProTip!
Adding no:label will show everything without a label.