Pull requests: NVIDIA/TensorRT-LLM
Pull requests list
[None][fix] Fix KVCacheV2Scheduler PEFT adapter ownership leak on suspend/resume (#18407)
#18412
opened Aug 30, 2026 by
Greninja44
Loading…
5 tasks done
[#17910][fix] Prevent compounding prefix when unpickling EngineDeadError
#18411
opened Aug 30, 2026 by
Ultron09
Loading…
4 tasks done
[None][feat] GVR V2 hint-free mode: pre_idx=None derives the top-k bracket from the current row
#18410
opened Aug 30, 2026 by
longcheng-nv
Collaborator
•
Draft
2 of 3 tasks
[TRTLLM-14959][refactor] declare MoE backend capabilities instead of checking exact classes
#18409
opened Aug 30, 2026 by
xxi-nv
Collaborator
Loading…
1 task done
[None][test] key perf-sanity case identity on test case name
#18408
opened Aug 30, 2026 by
chenfeiz0326
Collaborator
Loading…
7 tasks done
[None][test] add a basic disaggregated multi-node tier
#18404
opened Aug 29, 2026 by
fredricz-20070104
Collaborator
Loading…
4 tasks done
[https://nvbugs/6686534][fix] Kept the payload byte-identical (patch-id f393732542… on both sides) and…
#18403
opened Aug 29, 2026 by
trtllm-agent
Collaborator
Loading…
2 tasks done
[None][chore] Bump version to 1.3.0rc26
#18402
opened Aug 29, 2026 by
yuanjingx87
Collaborator
Loading…
1 task done
[https://nvbugs/6676511][fix] Reject unsupported speculative outputs
#18401
opened Aug 29, 2026 by
2ez4bz
Collaborator
Loading…
1 task done
[None][feat] Support num_postprocess_workers > 0 under the Ray Orchestration
ci: full pre-merge approved
#18399
opened Aug 29, 2026 by
shikicloud
Collaborator
Loading…
1 task done
[None][feat] Router Replay (R3): return per-token MoE routing
api-compatible
Accepted LLM API contract change that is backwards-compatible
#18397
opened Aug 29, 2026 by
shikicloud
Collaborator
Loading…
1 task done
[TRTLLM-10657][fix] Resolve MIXED_PRECISION quant config for DeepSeek W4A8 MoE experts
#18393
opened Aug 28, 2026 by
brnguyen2
Collaborator
Loading…
1 task done
[TRTLLM-14620][feat] Add confidence-guided dynamic verification for DeepSeek-V4 DSpark
#18392
opened Aug 28, 2026 by
zhaoyuanh-nvidia
Collaborator
•
Draft
1 task
[TRTLLM-14575][perf] Batch DSA cross-layer index remap into one kernel launch per indexer group
#18391
opened Aug 28, 2026 by
xwang233
Collaborator
Loading…
3 tasks done
[None][fix] Fix Qwen3.8-27B NVFP4 checkpoint loading
#18387
opened Aug 28, 2026 by
jupiterepoch
Loading…
1 task done
[https://nvbugs/6655990][test] Gate WAN multi-GPU LPIPS against a within-build reference and unwaive
VisualGen
#18386
opened Aug 28, 2026 by
chang-l
Collaborator
Loading…
1 task done
[https://nvbugs/6655986][test] Raise LTX-2 LPIPS golden thresholds to 0.15 and unwaive
VisualGen
#18384
opened Aug 28, 2026 by
chang-l
Collaborator
Loading…
1 task done
[TRTLLM-14575][perf] Use native next_n for DSA paged-MQA on Blackwell
#18383
opened Aug 28, 2026 by
xwang233
Collaborator
Loading…
1 task done
[https://nvbugs/6656598][fix] Fix Kimi-K2 E2E test
#18382
opened Aug 28, 2026 by
2ez4bz
Collaborator
Loading…
1 task done
[None][feat] Support image input in the Triton llmapi backend
#18381
opened Aug 28, 2026 by
faradawn
Collaborator
Loading…
[None][doc] Fix broken links across example and test READMEs
#18379
opened Aug 28, 2026 by
jlonge4
Loading…
1 task done
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.