Pull requests: NVIDIA/TensorRT-LLM
Pull requests list
[None][fix] Keep CuTe-DSL MLA decode for Kimi K3 H=96 speculative-verify batches
#18226
opened Aug 26, 2026 by
moraxu
Collaborator
Loading…
1 task done
[None][test] Remove deepseek v32 test cases on the qa side for disagg multinode perf testing
#18225
opened Aug 26, 2026 by
fredricz-20070104
Collaborator
Loading…
[None][infra] Waive 1 failed cases for main in pre-merge 56631
#18224
opened Aug 26, 2026 by
trtllm-agent
Collaborator
Loading…
[TRTLLM-15405][refactor] BREAKING: Remove TRTLLMSampler
api-breaking
Accepted LLM API contract change that is backwards-incompatible
#18223
opened Aug 26, 2026 by
zhaoyangwang-nvidia
Collaborator
•
Draft
1 task done
[None][test] Give Rubin its own perf block and stop it inheriting others
#18222
opened Aug 26, 2026 by
ruodil
Collaborator
Loading…
1 task done
[https://nvbugs/6337224][test] Unwaive test_config_database_tests_sync
#18221
opened Aug 26, 2026 by
hyukn
Collaborator
Loading…
[https://nvbugs/6620324][fix] Judge perf-runner liveness on any output, not on newlines
#18220
opened Aug 26, 2026 by
hyukn
Collaborator
Loading…
[https://nvbugs/6647310][chore] Unwaive TestLagunaXS::test_nvfp4
#18219
opened Aug 26, 2026 by
sunnyqgg
Collaborator
Loading…
1 task done
[TRTLLM-15376][test] expand Gemma 4 MTP QA coverage
#18218
opened Aug 26, 2026 by
jieli-matrix
Collaborator
Loading…
1 task done
[TRTLLMINF-263][infra] pass nSpect release entitlement
#18217
opened Aug 26, 2026 by
niukuo
Collaborator
Loading…
1 task done
DON'T REVIEW: LTX-2.3 visual generation support demo
VisualGen
#18215
opened Aug 26, 2026 by
yibinl-nvidia
Collaborator
•
Draft
1 task
[None][doc] Document the three generation-side time breakdown segments
#18213
opened Aug 25, 2026 by
nv-yna
Collaborator
Loading…
4 tasks done
[TRTLLMINF-339][infra] Enable BOLT profile-gen producer (job + cadence + promote)
ci: full pre-merge approved
#18212
opened Aug 25, 2026 by
mlefeb01
Collaborator
Loading…
1 task done
[TRTLLM-14388][refactor] BREAKING: Remove eagle_choices
api-breaking
Accepted LLM API contract change that is backwards-incompatible
#18210
opened Aug 25, 2026 by
mikeiovine
Collaborator
Loading…
1 task done
[https://nvbugs/6661914][fix] restore Wan 5B per-token AdaLN with TeaCache
VisualGen
#18209
opened Aug 25, 2026 by
chang-l
Collaborator
Loading…
[None][infra] Set --platform when retagging CI image
#18208
opened Aug 25, 2026 by
yuanjingx87
Collaborator
Loading…
1 task
[TRTLLM-15262][feat] Test CUDA graph input buffers status before replay
#18207
opened Aug 25, 2026 by
asfiyab-nvidia
Contributor
Loading…
1 task done
[None][perf] Fuse MiniMax-M3 QKV and index projection
api-compatible
Accepted LLM API contract change that is backwards-compatible
#18205
opened Aug 25, 2026 by
peihu-nv
Collaborator
Loading…
1 task done
[https://nvbugs/6644645][fix] Move the four entries into the pre-existing
optional_inplace_infos dict in…
#18204
opened Aug 25, 2026 by
trtllm-agent
Collaborator
Loading…
2 tasks done
[None][test] Unwaive test_trtllm_bench_llmapi_launch for nvbugs/6568058
ci: full pre-merge approved
#18203
opened Aug 25, 2026 by
allisonlim-nv
Contributor
Loading…
9 tasks done
[None][fix] register beneficial-to-skip contributions only after scheduling
#18202
opened Aug 25, 2026 by
eopXD
Collaborator
Loading…
1 task done
[https://nvbugs/6647349][fix] Replace FuzzyWuzzy with RapidFuzz
#18201
opened Aug 25, 2026 by
lfr-0531
Collaborator
Loading…
1 task done
[TRTLLM-15314][fix] recalculate PEFT cache capacity for adapter dtype
#18200
opened Aug 25, 2026 by
achartier
Collaborator
Loading…
1 task done
ProTip!
Mix and match filters to narrow down what you’re looking for.