Skip to content

Pull requests: NVIDIA/TensorRT-LLM

New pull request New
293 Open 2,460 Closed
293 Open 2,460 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Draft: test: [CI] Add failed cases into waives.txt
#5569 opened Jun 28, 2025 by xinhe-nv • Draft
[nvbug/5354946][fix] Fix mtp vanilla draft inputs
#5568 opened Jun 28, 2025 by lfr-0531 Loading…
6
Refactor the topk parallelization part for the routing kernels
#5567 opened Jun 28, 2025 by ChristinaZ Loading…
6
[feat][test] reuse MPI pool executor across tests
#5566 opened Jun 28, 2025 by omera-nv Loading…
11
Investigate Gemma3 1B discrepancy
#5564 opened Jun 28, 2025 by brb-nv • Draft
Fix GEMM+AR fusion on blackwell
#5563 opened Jun 28, 2025 by xavier-nvidia Loading…
test: Deprecate gpt_model_type "v1" static batching from triton_backe…
#5562 opened Jun 27, 2025 by mc-nv Loading…
Implement --served_model_name and improve command line parsing
#5561 opened Jun 27, 2025 by pathorn Loading…
[TRTLLM-4926][feat] Reimplement metrics endpoint with stats about requests
#5560 opened Jun 27, 2025 by pathorn Loading…
[fix] Use decorator for request cancelation and handle CancelledError
#5559 opened Jun 27, 2025 by pathorn Loading…
[nvbug/5337601][fix] Fix disagg + speculative decoding
#5558 opened Jun 27, 2025 by Tabrizian Loading…
1
Refactor moe permute and finalize op by removing duplicated code
#5557 opened Jun 27, 2025 by limin2021 Loading…
6
[TRTLLM-6104] feat: add request_perf_metrics to triton LLMAPI backend
#5554 opened Jun 27, 2025 by xuanzic Loading…
1
[feat] Support MXFP4 x BF16 Grouped GEMM in FusedMoE Pytorch Module
#5552 opened Jun 27, 2025 by jinyangyuan-nvidia Loading…
@StudyingShao @jinyangyuan-nvidia
8
feat: Improve dev container tagging
#5551 opened Jun 27, 2025 by ixlmar • Draft
3
feat: Optimize TRTLLM Sampler perf single beam single step
#5550 opened Jun 27, 2025 by dcampora Loading…
9
tests: add test_chunked_prefill for llama4
#5549 opened Jun 27, 2025 by xinhe-nv • Draft
Deduplicate waive list
#5546 opened Jun 27, 2025 by yiqingy0 Loading…
1
test: Use multiple workers for multi-GPU engine building
#5545 opened Jun 27, 2025 by Funatiq • Draft
5
rcca: test default kv_cache_reuse option for pytorch multimodal
#5544 opened Jun 27, 2025 by StanleySun639 Loading…
8
[draft] chore: enhance GenerationExecutor with RPC
#5543 opened Jun 27, 2025 by Superjomn • Draft
[DON'T MERGE] NGram with iter_stats
#5542 opened Jun 27, 2025 by wili-65535 • Draft
6
[nvbug 5304752][fix]: enhance _check_arguments to filter illegal requests for pytorch backend
#5541 opened Jun 27, 2025 by LinPoly Loading…
7
Add pd dynamic scaling readme
#5540 opened Jun 27, 2025 by Shunkangz Loading…
@Shunkangz
[Infra][release/0.21]Update nccl to 2.27.5
#5539 opened Jun 27, 2025 by EmmaQiaoCh Loading…
19
ProTip! Exclude everything labeled bug with -label:bug.