Skip to content

Pull requests: NVIDIA/TensorRT-LLM

New pull request New
267 Open 2,065 Closed
267 Open 2,065 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

[feat]: improve performance of XQA-MLA for sm120
#5087 opened Jun 10, 2025 by lowsfer Loading…
[feat] Support torch compile for attention dp
#5086 opened Jun 10, 2025 by liji-nv Loading…
2
test: strictly constraint disaggregated serving llama4 to H200
#5085 opened Jun 10, 2025 by StanleySun639 Loading…
2
[fix] Fix test_attention_mla
#5084 opened Jun 10, 2025 by jinyangyuan-nvidia Loading…
@jinyangyuan-nvidia
2
test: add more cases for rtx_pro_6000_se and add option kv_cache_dtype in perf test
#5083 opened Jun 10, 2025 by ruodil Loading…
chore: Mass integration of release/0.20
#5082 opened Jun 10, 2025 by amirkl94 Loading…
3
ci: waive test [NVBUGS/5301492]
#5081 opened Jun 10, 2025 by stnie Loading…
Fix LoRA broadcast issue in pytorch flow
#5080 opened Jun 10, 2025 by amitz-nv • Draft
2
refactor: improve code readability
#5079 opened Jun 10, 2025 by Shixiaowei02 Loading…
11
fix: fix cuda graph max batch size for spec decoding cases.
#5076 opened Jun 10, 2025 by lfr-0531 Loading…
2
infra[TRTLLM-5635] remove package stage in CI build
#5075 opened Jun 10, 2025 by niukuo Loading…
2
[fix]: Fall back to HMAC to Avoid IPC Serialization Churn
#5074 opened Jun 10, 2025 by yibinl-nvidia • Draft
1
[TRTLLM-5786][test] Add llama3.2-3b test case
#5073 opened Jun 10, 2025 by crazydemo Loading…
3
fix: remove duplicate trust_remote_code from serve command
#5072 opened Jun 10, 2025 by yechank-nvidia Loading…
@yechank-nvidia
2
Draft: test: skip disaggregated tests on arm
#5070 opened Jun 10, 2025 by xinhe-nv • Draft
chore: rename IOFormatter to BaseCacheFormatter
#5068 opened Jun 10, 2025 by zhengd-nv Loading…
3
test(perf): Add remaining Llama-Nemotron perftests (nano, super, ultra) + extras ✨
#5066 opened Jun 10, 2025 by venkywonka Loading…
4 tasks done
13
[https://nvbugspro.nvidia.com/bug/5332927][fix] Fix the bug in the routing unit test
#5065 opened Jun 10, 2025 by ChristinaZ Loading…
5
Draft: test: [CI] remove closed bugs
#5063 opened Jun 10, 2025 by xinhe-nv • Draft
Fix:Add decodingcondfig into serialization
#5062 opened Jun 10, 2025 by nv-guomingz • Draft
chore: build the wheel using NIXL as the default
#5061 opened Jun 10, 2025 by Shixiaowei02 Loading…
3
bugfix [AutoDeploy]: Correct usage of pytorch_config in autodeploy integration of trtllm-bench
#5059 opened Jun 10, 2025 by suyoggupta Loading…
1
[https://nvbugs/5277592][fix] fix cuda graph padding for spec decoding (only for 0.20)
#5058 opened Jun 10, 2025 by lfr-0531 Loading…
2
test(perf): Add Llama-3_1-Nemotron-Ultra-253B-v1 perf tests (pyt, fp8)
#5057 opened Jun 10, 2025 by venkywonka • Draft
None: fix OOM because of unnecessary mha workspace
#5056 opened Jun 10, 2025 by ttyio Loading…
11
ProTip! Type g i on any issue or pull request to go back to the issue listing page.