Skip to content

Pull requests: vllm-project/vllm

New pull request New
1,290 Open 16,480 Closed
1,290 Open 16,480 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

[Quantization] enable compressed-tensors marlin support for turing (2)
#31008 opened Dec 19, 2025 by jinzhen-lin Loading…
[Qwen3-Omni] fixed _get_feat_extract_output_lengths function qwen Related to Qwen models
#31007 opened Dec 19, 2025 by wangxiongts Loading…
5 tasks
3
[Bugfix] Fix Ray GPU availability warning message v1
#31006 opened Dec 19, 2025 by jarieshan Loading…
5 tasks
4
[Mics] add pcp basic support to MoE model
#31003 opened Dec 19, 2025 by pisceskkk Loading…
1
feat(kernel): patch fused_gdn_gating qwen Related to Qwen models
#31002 opened Dec 19, 2025 by OsirisDuan Loading…
5 tasks
3
`Feature/nvfp4 universal fallback emulation
#31001 opened Dec 19, 2025 by Rob-P-Smith • Draft
3
[Perf][ROCm][AWQ] Improve performance of fused MoE GPTQ-AWQ and AWQ dequant kernels rocm Related to AMD ROCm
#30998 opened Dec 19, 2025 by yuttian1 Loading…
6
Add Molmo2 multimodal model support documentation Improvements or additions to documentation multi-modality Related to multi-modality (#4194) new-model Requests to new models
#30997 opened Dec 19, 2025 by sangho-vision Loading…
4 of 5 tasks
29
[WIP]improve cpu Benchmark Suite tests for 0.12.0 ci/build cpu Related to CPU backends performance Performance-related issues
#30994 opened Dec 19, 2025 by louie-tsai Loading…
5 tasks
4
Bump Flashinfer to v0.6.0rc1 ci/build nvidia
#30993 opened Dec 18, 2025 by elvischenv Loading…
5 tasks
[Misc] Remove deprecated metric vllm:time_per_output_token_seconds for v0.13 release v1
#30992 opened Dec 18, 2025 by jliu9515 Loading…
3 of 5 tasks
3
[ROCm][CI/Build] Update ROCm dockerfiles ci/build rocm Related to AMD ROCm
#30991 opened Dec 18, 2025 by gshtras Loading…
4
[MoE Refactor][3/N] Deprecate cutlass block quant fp8 (b200) nvidia ready ONLY add when PR is ready to merge/full CI is needed
#30990 opened Dec 18, 2025 by robertgshaw2-redhat Loading…
5 tasks
8
[P/D] Add utility to slow down server for decode benchmarking frontend kv-connector v1
#30988 opened Dec 18, 2025 by hjjq • Draft
2
Eagle 3 fix sometimes when you don't set architectures you get "Model architectures ['EagleLlamaModel'] are not supported for now. llama Related to Llama models new-model Requests to new models
#30987 opened Dec 18, 2025 by aidando73 Loading…
5 tasks
7
Grid construction based on num_active_lora and support CUDA graph capture across various num_active_lora nvidia v1
#30984 opened Dec 18, 2025 by yugong333 Loading…
5 tasks
5
Update Pytorch version update docs documentation Improvements or additions to documentation
#30982 opened Dec 18, 2025 by atalman Loading…
1
blackwell frontend v1
#30981 opened Dec 18, 2025 by dtunikov Loading…
7
[Do not merge][Async] Asynchronous DP coordination v1
#30980 opened Dec 18, 2025 by MatthewBonanni • Draft
3 of 5 tasks
5
[MoE Refactor] Add mk for cutlass fp8 block nvidia v1
#30979 opened Dec 18, 2025 by robertgshaw2-redhat • Draft
4
Add positional embedding and kv_cache fusion for llama and gpt-oss gpt-oss Related to GPT-OSS models llama Related to Llama models v1
#30978 opened Dec 18, 2025 by dllehr-amd • Draft
5 tasks
4
Docs: add OpenAI SDK example for Qwen2.5-VL classification documentation Improvements or additions to documentation qwen Related to Qwen models
#30977 opened Dec 18, 2025 by Dhruv-80 Loading…
1
7
Use aiter triton fused_add_rmsnorm_pad for gpt-oss gpt-oss Related to GPT-OSS models
#30976 opened Dec 18, 2025 by Rohan138 • Draft
5 tasks
2
[Misc] Disable default --ready-check-timeout-sec extra call in vllm bench performance Performance-related issues
#30975 opened Dec 18, 2025 by NickLucche Loading…
[Bugfix] Fix incorrect tiles creation for mm prefix triton attention ready ONLY add when PR is ready to merge/full CI is needed
#30974 opened Dec 18, 2025 by Isotr0py Loading…
5 tasks
5
ProTip! Adding no:label will show everything without a label.