Pull requests: NVIDIA/TensorRT-LLM
Pull requests list
[#19463][feat] Add optional ModelScope model loading support
#19464
opened Sep 21, 2026 by
functionstackx
Loading…
3 of 8 tasks
[None][ci] waive pre-existing test failures on main
#19462
opened Sep 20, 2026 by
brnguyen2
Collaborator
Loading…
[#19460][feat] Expose image, video, and audio prompt-token counts in trtllm-serve responses
#19461
opened Sep 20, 2026 by
Ritabanm
Loading…
4 tasks done
[None][feat] OpenAI protocol: cap tool enum values and accept parallel_tool_calls
api-compatible
Accepted LLM API contract change that is backwards-compatible
#19459
opened Sep 20, 2026 by
brnguyen2
Collaborator
Loading…
1 task done
[None][fix] Validate media data-URIs and gate private-URL fetches behind an opt-in
#19458
opened Sep 20, 2026 by
brnguyen2
Collaborator
Loading…
1 task done
[None][fix] Report context-length-exceeded with OpenAI-style message and error code
#19457
opened Sep 20, 2026 by
brnguyen2
Collaborator
Loading…
1 task done
[None][fix] Tear down cleanly when multi-rank warmup fails
#19456
opened Sep 20, 2026 by
brnguyen2
Collaborator
Loading…
1 task done
[None][fix] Drain guided-decoder host functions before GIL-holding forward calls
#19455
opened Sep 20, 2026 by
brnguyen2
Collaborator
Loading…
1 task done
[None][fix] Responses API: treat explicit null as unset for optional fields
#19454
opened Sep 20, 2026 by
brnguyen2
Collaborator
Loading…
1 task done
[None][doc] Make the PR title examples pass the PR title check
#19452
opened Sep 20, 2026 by
100milliongold
Contributor
Loading…
[#18609][fix] Do not require encoder tokens for cross prefix reuse
#19450
opened Sep 20, 2026 by
100milliongold
Contributor
Loading…
[https://nvbugs/6786567][chore] Remove waiver for SM103 GVR CUDA graph test
#19449
opened Sep 20, 2026 by
longcheng-nv
Collaborator
Loading…
1 task done
[#19362][fix] Name the act fusion in the skip reason
#19448
opened Sep 20, 2026 by
100milliongold
Contributor
Loading…
[https://nvbugs/6713426][chore] Gate and instrument the MegaMoE DeepGEMM SymmBuffer rendezvous
#19447
opened Sep 20, 2026 by
xxi-nv
Collaborator
Loading…
[None][perf] PDL on the missing decode edges
#19446
opened Sep 20, 2026 by
brb-nv
Collaborator
Loading…
1 task done
[None][build]] Improve C++ build structure
#19445
opened Sep 19, 2026 by
Funatiq
Collaborator
Loading…
1 task
[None][fix] Synchronize disaggregated KV receive destinations
#19444
opened Sep 19, 2026 by
xwang233
Collaborator
Loading…
[https://nvbugs/6778910][fix] Include retiring requests in ADP router placement load view
#19443
opened Sep 19, 2026 by
trtllm-agent
Collaborator
Loading…
2 tasks done
[None][feat] support W4A16 GPTQ in the PyTorch backend
#19442
opened Sep 19, 2026 by
zupengwang
Loading…
Feat/unified dspark kv cache disagg
#19441
opened Sep 19, 2026 by
allisonlim-nv
Contributor
•
Draft
1 task
[None][feat] Support model_kwargs with encoder CUDA graphs
#19439
opened Sep 19, 2026 by
tingyangk
Collaborator
Loading…
1 task done
[#19435][fix] Keep the insufficient-KV vote rank-safe under attention-DP
#19436
opened Sep 18, 2026 by
0z5a
Loading…
ProTip!
Mix and match filters to narrow down what you’re looking for.