Skip to content

Pull requests: vllm-project/vllm

New pull request New
3,452 Open 26,764 Closed
3,452 Open 26,764 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

[XPU][Attention] Enable tensor-descriptor Q load for non-power-of-2 GQA (~1.9x Qwen2-7B) intel-gpu Related to Intel GPU qwen Related to Qwen models v1
#47248 opened Jul 1, 2026 by gurwinderintel Loading…
1
[Bugfix][Frontend] Include usage in non-streaming GenerateResponse bug Something isn't working frontend
#47247 opened Jul 1, 2026 by aoshen02 Collaborator Loading…
4 tasks
[XPU][Attention] Widen prefill BLOCK_M to 32 for power-of-2 GQA (~2x prefill on BMG) intel-gpu Related to Intel GPU v1
#47246 opened Jul 1, 2026 by gurwinderintel Loading…
1
add sycl path for Mhc
#47245 opened Jul 1, 2026 by xiaolong-intel • Draft
4 tasks
1
[Core] Make sleep-mode backend capability flags communicator-agnostic ready ONLY add when PR is ready to merge/full CI is needed v1
#47243 opened Jul 1, 2026 by matteso1 Contributor Loading…
1
[CI/Build] Fix LoRA testing ready ONLY add when PR is ready to merge/full CI is needed
#47242 opened Jul 1, 2026 by jeejeelee Member Loading…
4 tasks
4
[do not merge][test only][xpu][ci]Debug Intel B50 agent ci/build intel-gpu Related to Intel GPU
#47240 opened Jul 1, 2026 by zxd1997066 Contributor • Draft
4 tasks
[BugFix][Spec Decode] Compact shared topk indices buffer after first MTP draft step bug Something isn't working deepseek Related to DeepSeek models ready ONLY add when PR is ready to merge/full CI is needed speculative-decoding v1
#47238 opened Jul 1, 2026 by TheEpicDolphin Collaborator Loading…
1
[Bugifx][INC] Fix INC quantization method selection for non-quantized layers
#47237 opened Jul 1, 2026 by lvliang-intel Loading…
4 tasks
4
fix(simple_kv_offload): restore evicted CPU blocks and fix cursor advancement v1
#47235 opened Jul 1, 2026 by Alex-ai-future Contributor Loading…
1
[KVOffload] Document CPU eviction behavior as reference (control branch) v1
#47234 opened Jul 1, 2026 by Alex-ai-future Contributor Loading…
Revert "Remove more unnecessary load_weights methods" (#47058) deepseek Related to DeepSeek models llama Related to Llama models mistral Related to Mistral models multi-modality Related to multi-modality (#4194) qwen Related to Qwen models speculative-decoding
#47233 opened Jul 1, 2026 by vllm-agent Contributor • Draft
2
Revert "[Platform] Replace torch.cuda.mem_get_info with torch.accelerator.get_memory_info" (#44825) cpu Related to CPU backends intel-gpu Related to Intel GPU kv-connector multi-modality Related to multi-modality (#4194) nvidia v1
#47232 opened Jul 1, 2026 by vllm-agent Contributor • Draft
[XPU][CI] Add tests/v1/e2e/general/test_correctness_sliding_window.py in Intel GPU CI ci/build intel-gpu Related to Intel GPU
#47231 opened Jul 1, 2026 by zxd1997066 Contributor Loading…
4 tasks done
fix(serve): return HTTP 422 instead of 500 for image/media URL fetch errors frontend gpt-oss Related to GPT-OSS models multi-modality Related to multi-modality (#4194)
#47230 opened Jul 1, 2026 by aoright Loading…
1
4
[DSV4] Better MXFP8 quantization kernel ready ONLY add when PR is ready to merge/full CI is needed
#47229 opened Jul 1, 2026 by zyongye Member Loading…
1
fix(deepseek_v4): resolve auto kv-cache-dtype to fp8_ds_mla on SM120 deepseek Related to DeepSeek models nvidia
#47228 opened Jul 1, 2026 by hclsys Contributor Loading…
1
[Doc] add AI Runway to integrations documentation Improvements or additions to documentation
#47227 opened Jul 1, 2026 by robert-cronin Loading…
4 tasks done
2
[LoRA] Integrate flashinfer MoE LoRA nvidia
#47226 opened Jul 1, 2026 by jeejeelee Member • Draft
4 tasks
[Perf][Model] Build Phi4MM Conformer streaming mask on target device
#47225 opened Jun 30, 2026 by Juice-XIJ Loading…
4 tasks done
1
[Perf] Build prompt mask with presence-only scatter in apply_penalties
#47224 opened Jun 30, 2026 by Arthur-St-06 Loading…
1
[Bugfix] Fix online-quant MoE loading zero weights after #47058 bug Something isn't working
#47221 opened Jun 30, 2026 by mgoin Member Loading…
1
[AMD][EPLB] Enable EPLB for Quark OCP MXFP4 MoE ready ONLY add when PR is ready to merge/full CI is needed rocm Related to AMD ROCm
#47220 opened Jun 30, 2026 by okorzh-amd Contributor Loading…
4 tasks
2
[Bugfix] Include SM12x in _is_fa4_supported() compute capability check bug Something isn't working
#47218 opened Jun 30, 2026 by tgmerritt Loading…
2
[Bugfix][Gemma4] Keep image bidirectional attention within the sliding window bug Something isn't working v1
#47217 opened Jun 30, 2026 by lucianommartins Contributor Loading…
3
ProTip! Mix and match filters to narrow down what you’re looking for.