Skip to content

Pull requests: NVIDIA/TensorRT-LLM

New pull request New
380 Open 4,587 Closed
380 Open 4,587 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

[TRTLLM-8269][test] do not explicitly pass temperature=0 to select greedy sampling
#8110 opened Oct 1, 2025 by ixlmar • Draft
1 task done
2
[TRTLLM-8160][TRTLLM-8161][feat][draft]Add runtime logic for draft token tree
#8109 opened Oct 1, 2025 by yweng0828 • Draft
1 task
[None][fix] AutoDeploy: dive deeper into token generation bugs AutoDeploy <NV> AutoDeploy Backend
#8108 opened Oct 1, 2025 by lucaslie • Draft
1 task done
@lucaslie
1
Draft: [None][fix] fix: updating patchelf version
#8105 opened Sep 30, 2025 by pcastonguay Loading…
1 task
4
[TRTLLM-5966][feat] Helix: add full MLA support for Helix
#8104 opened Sep 30, 2025 by MatthiasKohl Loading…
14
[doc] Add Qwen3 Next Guide to Core README Community want to contribute PRs initiated from Community
#8101 opened Sep 30, 2025 by faradawn Loading…
1 task
5
[https://nvbugs/5521949][fix] Fix head_size handling in ModelConfig.get_bindings_model_config
#8100 opened Sep 30, 2025 by amitz-nv Loading…
1 task
@amitz-nv
7
[#7588][feat] lock gpu clocks in test_perf.py to reliably detect perf regressions
#8099 opened Sep 30, 2025 by MrGeva Loading…
1 task done
12
[https://nvbugs/5541494] [fix] Fix missing sm100f/103a kernels and add tests
#8098 opened Sep 30, 2025 by VALLIS-NERIA Loading…
1 task
3
[None][feat] reuse cudagraph memory pool in normal forward flow
#8095 opened Sep 30, 2025 by HuiGao-NV Loading…
1 task
12
[None][fix] Avoid unnecessary concat in attn_output_gate case.
#8094 opened Sep 30, 2025 by yuxianq Loading…
1 task done
10
[https://nvbugs/5550722][fix] Fix image load
#8093 opened Sep 30, 2025 by yechank-nvidia Loading…
@yechank-nvidia
6
[#7588][fix] fixed the kv cache size parsing in test_perf.py AD backend
#8092 opened Sep 30, 2025 by MrGeva Loading…
1 task done
@MrGeva
18
[TRTLLM-8246][test] add multimodal kvcache+chunked_prefil cases in to QA test list
#8091 opened Sep 30, 2025 by crazydemo Loading…
1 task done
4
[None][fix] Disable DeepGEMM for Qwen3 MoE Attention layers
#8087 opened Sep 30, 2025 by achartier Loading…
1 task done
4
[None][feat] add RocketKV support (experimental)
#8086 opened Sep 30, 2025 by lfr-0531 Loading…
1 task
45
[None][fix] Add Lock to protect mReqeustToSession
#8085 opened Sep 30, 2025 by chuangz0 Loading…
1 task done
5
[None][feat] Spark dev branch 1.1rc3 spark gpt oss with user look up table not to be merged
#8084 opened Sep 30, 2025 by farazkh80 • Draft
1 task done
6
[None][fix] Enable FP8 ContextMLA on GB300
#8080 opened Sep 30, 2025 by longlee0622 Loading…
1 task done
@longlee0622
10
[None][autodeploy] small refactors on attention matching
#8079 opened Sep 30, 2025 by Fridah-nv Loading…
1 task done
@Fridah-nv
10
[https://nvbugs/5549111][fix] Fix 2-model overlap scheduler accuracy on very long prompts
#8076 opened Sep 29, 2025 by mikeiovine Loading…
1 task done
18
[None][fix] Fix Qwen3 FP8 per-tensor when requesting TRTLLM-GEN MoE backend
#8075 opened Sep 29, 2025 by achartier Loading…
1 task done
17
test gb200
#8074 opened Sep 29, 2025 by yuanjingx87 • Draft
1 task
9
[#7312][feat] Torch.compile for transformers mode
#8073 opened Sep 29, 2025 by h-guo18 • Draft
1 task
[https://nvbugs/5549081][fix] Fix device id assignment for some vision models
#8070 opened Sep 29, 2025 by chang-l Loading…
1 task done
10
ProTip! Type g i on any issue or pull request to go back to the issue listing page.