Pull requests: NVIDIA/TensorRT-LLM
Pull requests list
[TRTLLM-11759][fix] Reduce peak host memory during NemotronH_Nano_VL_V2 init
#13283
opened Apr 21, 2026 by
pamelap-nvidia
Collaborator
•
Draft
1 task done
[https://nvbugs/6098442][fix] Update fmha attention cubins
#13282
opened Apr 21, 2026 by
heyuhhh
Collaborator
Loading…
1 task done
[https://nvbugs/5973199][fix] Add NCCL fallback for AutoDeploy MoE alltoall when MNNVL is unavailable
#13281
opened Apr 21, 2026 by
tensorrt-cicd
Collaborator
Loading…
2 tasks done
[None][cleanup] remove legacy addSequence path
#13280
opened Apr 21, 2026 by
liji-nv
Collaborator
Loading…
1 task done
[https://nvbugs/6087946][fix] TestGemma3_1BInstruct::test_fp8_prequantized missing @skip_pre_ada decorator, ca
#13279
opened Apr 21, 2026 by
tensorrt-cicd
Collaborator
Loading…
2 tasks done
[None][fix] Fix Mamba cache correctness under MTP + CUDA-graph padding
#13278
opened Apr 21, 2026 by
Wanli-Jiang
Collaborator
•
Draft
1 task done
[https://nvbugs/6075556][fix] ** During PP memory profiling warmup,
_executor_loop_cleanup() blocks indefini
#13277
opened Apr 21, 2026 by
tensorrt-cicd
Collaborator
Loading…
2 tasks done
[TRI-966] [fix] Fix L0_backend_trtllm
Community want to contribute
PRs initiated from Community
#13276
opened Apr 21, 2026 by
pskiran1
Contributor
Loading…
1 task done
[None][fix] initialize sampler state for ADP dummy requests
#13275
opened Apr 21, 2026 by
bobboli
Collaborator
Loading…
[https://nvbugs/6075345][fix] test_llmapi_launch_multiple_tasks ignored the
task_script parameter and always
#13273
opened Apr 21, 2026 by
tensorrt-cicd
Collaborator
Loading…
2 tasks done
[TRTLLM-13271][feat] Artifact cleanup
#13272
opened Apr 21, 2026 by
greg-kwasniewski1
Collaborator
Loading…
4 tasks done
[https://nvbugs/6079919][fix] In MTPEagleWorker's first MTP sub-step, block_ids_per_seq was reordered (generat
#13270
opened Apr 21, 2026 by
tensorrt-cicd
Collaborator
Loading…
2 tasks done
[https://nvbugs/6078421][fix] Commit dcb4a71b3 intentionally disabled
initialize_mrope_delta_cache in `qwen3
#13269
opened Apr 21, 2026 by
tensorrt-cicd
Collaborator
Loading…
2 tasks done
[https://nvbugs/6095953][fix] Fix cache memory estimation for Qwen3 hybrid models in trtllm-bench
#13268
opened Apr 21, 2026 by
hyukn
Collaborator
Loading…
1 task done
[https://nvbugs/6095421][fix] fix PP>=3 executor shutdown hang in broadcast sample state loop
#13267
opened Apr 21, 2026 by
yihwang-nv
Collaborator
Loading…
[None][test] Add disagg multinode perf new cases
#13266
opened Apr 21, 2026 by
fredricz-20070104
Collaborator
Loading…
[None][chore] Adjust the current disagg cases
#13265
opened Apr 21, 2026 by
fredricz-20070104
Collaborator
Loading…
[https://nvbugs/6062537][fix]
/metrics endpoint served JSON iteration stats instead of Prometheus exposition
#13264
opened Apr 21, 2026 by
tensorrt-cicd
Collaborator
Loading…
2 tasks done
[https://nvbugs/6097980][fix] The
_visual_gen_deps fixture runs apt-get update and `apt-get install ffmpeg
#13263
opened Apr 21, 2026 by
tensorrt-cicd
Collaborator
Loading…
2 tasks done
[TRTLLM-12062][test] remove obsolete model tests
#13262
opened Apr 21, 2026 by
xinhe-nv
Collaborator
Loading…
1 task done
[None][chore] Add related trtllm-gen attention kernel files to trigger multi-gpu tests
#13260
opened Apr 21, 2026 by
heyuhhh
Collaborator
Loading…
1 task done
[https://nvbugs/6084445][fix] use DEEPGEMM for DeepSeek-V3-Lite fp8 chunked prefill on SM100/SM103
#13257
opened Apr 21, 2026 by
jmydurant
Collaborator
Loading…
1 task
[#13099][feat] Support Wan 2.2 5B TI2V model
#13256
opened Apr 21, 2026 by
abc99lr
Loading…
1 task done
ProTip!
Updated in the last three days: updated:>2026-04-18.