Skip to content

Pull requests: NVIDIA/TensorRT-LLM

New pull request New
124 Open 509 Closed
124 Open 509 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

test: add random image test for llama-3.2-11b-vision
#3055 opened Mar 25, 2025 by crazydemo Loading…
5
feat: Add EXAONE-Deep
#3054 opened Mar 25, 2025 by yechank-nvidia Loading…
@yechank-nvidia
4
fix: Set correct draft_token_nums to dummy requests for torch compilation with MTP
#3053 opened Mar 25, 2025 by HuiGao-NV Loading…
7
doc: Update DeepSeekV3 doc
#3052 opened Mar 25, 2025 by xiaoweiw-nv Loading…
10
chore: upgrade transformers to 4.50.0
#3051 opened Mar 25, 2025 by achartier Loading…
3
fix: AllReduce CUDA Graph Fix
#3049 opened Mar 25, 2025 by yizhang-nv • Draft
@yizhang-nv
feat: Support prequantized fp8 ckpt for nemotron-mini-4b-instruct
#3046 opened Mar 24, 2025 by brb-nv Loading…
4
Feat: Support Linear block scale layout in FP4 quantization
#3045 opened Mar 24, 2025 by yibinl-nvidia Loading…
@yibinl-nvidia
1
feat: Pytorch PP + attention DP support
#3044 opened Mar 24, 2025 by achartier Loading…
@achartier
7
chore: Add second possible output for llava
#3043 opened Mar 24, 2025 by amukkara Loading…
10
fix: [NVBUG 5087143] Fix vila test
#3042 opened Mar 24, 2025 by yuanjings-nvda • Draft
@yuanjings-nvda
7
perf: [AutoDeploy] Enable AutoDeploy as a backend in trtllm-bench
#3041 opened Mar 24, 2025 by suyoggupta Loading…
11
perf: Readd iteration logging for trtllm-bench.
#3039 opened Mar 24, 2025 by FrankD412 Loading…
@FrankD412 @hchings
6
chore: [TRTLLM-3694] Move functional args to llmargs
#3036 opened Mar 24, 2025 by hchings Loading…
@hchings
6
feat: Add initial EAGLE-3 implementation
#3035 opened Mar 24, 2025 by mikeiovine Loading…
@mikeiovine
1
feat: Unify two versions of allreduce custom op
#3032 opened Mar 24, 2025 by yukunh-nvidia Loading…
@yukunh-nvidia
infra: [CI] - Only checkout the Git sourcecodes once in the CI pipeline
#3029 opened Mar 24, 2025 by chzblych Loading…
11
feat: allocate minimal blocks per window size
#3028 opened Mar 24, 2025 by netanel-haber • Draft
test: [TRTLLM-4000] Port multi GPU changes to GitHub CI Any issue relates with CI testing
#3027 opened Mar 24, 2025 by DomBrown Loading…
@DomBrown
12
feat: Draft/lora_modules_support
#3026 opened Mar 24, 2025 by danielafrimi • Draft
chore: refactor the LlmArgs for better api reference doc and easier iterations
#3025 opened Mar 24, 2025 by Superjomn • Draft
refactor: Remove speculative decoding parameters from stateful decoders
#3024 opened Mar 24, 2025 by Funatiq Loading…
6
fix: disable KV cache reuse if using attention sink
#3021 opened Mar 24, 2025 by Funatiq Loading…
11
Support cos_sin_cache in all cases.
#3020 opened Mar 24, 2025 by yuxianq Loading…
5
fix: creating output of dataset generator in current directory bug Something isn't working
#3018 opened Mar 24, 2025 by hypdeb Loading…
@hypdeb
3
ProTip! Exclude everything labeled bug with -label:bug.