Pull requests: NVIDIA/TensorRT-LLM
Pull requests list
[NVBUG-6448152][test] isolate CTX consensus regression
#16572
opened Jul 19, 2026 by
chienchunhung
Collaborator
•
Draft
[TRTLLM-14417][fix] Exclude ADP/cuda-graph dummy requests from speculative-decode acceptance stats
#16571
opened Jul 19, 2026 by
xwang233
Collaborator
Loading…
3 tasks done
[None][perf] Fuse Triton FP8 dynamic-quant amax for dense projections
#16569
opened Jul 18, 2026 by
hyukn
Collaborator
Loading…
3 tasks done
[https://nvbugs/6442073][fix] Enforce a fixed max_seq_len for Qwen's AttentionOp caching purposes
#16568
opened Jul 18, 2026 by
moraxu
Collaborator
Loading…
1 task done
[NVBUG-6448152][test] trace disagg transfer lifecycle
#16567
opened Jul 18, 2026 by
chienchunhung
Collaborator
•
Draft
[NVBUG-6448152][test] test #15139 with sender wakeup fix
#16566
opened Jul 18, 2026 by
chienchunhung
Collaborator
•
Draft
[NVBUG-6448152][test] control #15139 with sender wakeup fix
#16565
opened Jul 18, 2026 by
chienchunhung
Collaborator
•
Draft
[fix][perf][sampler] Add batched-pybind fast-path in TorchSampler.update_requests for gpt-oss-120b
#16564
opened Jul 18, 2026 by
chenfeiz0326
Collaborator
Loading…
6 tasks done
[None][feat] Support the DMD2-distilled Cosmos3-Super-Text2Image-4Step checkpoint
VisualGen
#16563
opened Jul 18, 2026 by
ishovkun
Loading…
1 task done
[None][perf] Add selectable SafeTensors weight-load policies
#16562
opened Jul 18, 2026 by
chienchunhung
Collaborator
•
Draft
[None][infra] Waive 1 failed cases for main in pre-merge 48383
#16560
opened Jul 18, 2026 by
trtllm-agent
Collaborator
Loading…
[https://nvbugs/6312828][test] instrument disaggregated admission telemetry
#16559
opened Jul 17, 2026 by
chienchunhung
Collaborator
•
Draft
[TRTLLMINF-102][fix] Surface SLURM device faults to the failure classifier
#16557
opened Jul 17, 2026 by
dpitman-nvda
Collaborator
Loading…
1 task done
[NVBUG-6448152][test] measure exact post-#15139 merge treatment
#16556
opened Jul 17, 2026 by
chienchunhung
Collaborator
•
Draft
[NVBUG-6448152][test] measure exact pre-#15139 merge control
#16555
opened Jul 17, 2026 by
chienchunhung
Collaborator
•
Draft
[NVBUG-6448152][test] measure exact pre-#15139 baseline (DO NOT REVIEW/MERGE)
#16553
opened Jul 17, 2026 by
chienchunhung
Collaborator
•
Draft
[#16549][fix] Make gRPC GenerateRequest.max_tokens optional
#16552
opened Jul 17, 2026 by
tanmayv25
Collaborator
Loading…
[None][infra] Add per-stage opt to cap/disable stage-level infra retries
#16551
opened Jul 17, 2026 by
dpitman-nvda
Collaborator
Loading…
1 task done
[https://nvbugs/6432953][fix] Fix MPI world heap corruption during teardown
#16550
opened Jul 17, 2026 by
mikeiovine
Collaborator
Loading…
1 task done
[https://nvbugs/6451032][fix] Reserve extra dflash context slot for dummy requests
#16548
opened Jul 17, 2026 by
amukkara
Collaborator
Loading…
1 task done
[None][feat] Raise the CuTE-DSL top-k decode limit to 16384 and support odd top_k
#16546
opened Jul 17, 2026 by
Hudayday
Collaborator
Loading…
ProTip!
Follow long discussions with comments:>50.