Skip to content

Pull requests: vllm-project/vllm

New pull request New
801 Open 10,099 Closed
801 Open 10,099 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

[V0 deprecation] Deprecate V0 Neuron backend ci/build documentation Improvements or additions to documentation ready ONLY add when PR is ready to merge/full CI is needed speculative-decoding
#21159 opened Jul 18, 2025 by WoosukKwon Loading…
3
[WIP] Support relaxed acceptance for thinking tokens in speculative decoding speculative-decoding
#21157 opened Jul 18, 2025 by Ximingwang-09 • Draft
4 tasks
8
Add an optimization doc on TPU documentation Improvements or additions to documentation tpu Related to Google TPUs
#21155 opened Jul 18, 2025 by bvrockwell • Draft
4 tasks
7
Remove xformers requirement for Mistral-format Pixtral and Mistral3
#21154 opened Jul 18, 2025 by wenchen76 • Draft
1 of 4 tasks
1
3
[Attention] Add support for "splitting" the CommonAttentionMetadata v1
#21153 opened Jul 18, 2025 by SageMoore • Draft
4
[V0 Deprecation] Remove V0 Spec Decode workers ci/build new-model Requests to new models ready ONLY add when PR is ready to merge/full CI is needed rocm Related to AMD ROCm speculative-decoding
#21152 opened Jul 18, 2025 by WoosukKwon Loading…
2
Implement structural_tag and json_schema for non-chat completion frontend
#21150 opened Jul 17, 2025 by pathorn Loading…
3
[Bugfix] Capture ray import error as string to avoid persistent references
#21149 opened Jul 17, 2025 by tjohnson31415 Loading…
3
[Misc] Make MM embedding merge interface explicit in model runner ready ONLY add when PR is ready to merge/full CI is needed tpu Related to Google TPUs v1
#21147 opened Jul 17, 2025 by ywang96 Loading…
4 tasks
6
[Core] disable gc during cuda graph capture codex startup-ux v1
#21146 opened Jul 17, 2025 by mgoin Loading…
6
Enable multi-image support benchmarking for serving performance Performance-related issues
#21145 opened Jul 17, 2025 by leopck Loading…
7
[Misc] allow pulling vllm in Ray runtime environment
#21143 opened Jul 17, 2025 by eric-higgins-ai Loading…
@ruisearch42
2
[Core] Add request preprocess counter in v1 v1
#21139 opened Jul 17, 2025 by vladmihailescu Loading…
3 of 4 tasks
6
[Attention] Optimize FlashInfer MetadataBuilder Build call rocm Related to AMD ROCm speculative-decoding v1
#21137 opened Jul 17, 2025 by LucasWilkinson Loading…
3 of 4 tasks
4
[Perf] Using mul instead of div for int8 quant
#21136 opened Jul 17, 2025 by yewentao256 Loading…
6
[Misc] change default request logging behavior to off codex
#21135 opened Jul 17, 2025 by simon-mo Loading…
9
Convert tests to ruff-format deepseek Related to DeepSeek models llama Related to Llama models multi-modality Related to multi-modality (#4194) performance Performance-related issues qwen Related to Qwen models rocm Related to AMD ROCm speculative-decoding structured-output tool-calling v1
#21129 opened Jul 17, 2025 by hmellor Loading…
3
[Core] Set pooling params based on task and model frontend ready ONLY add when PR is ready to merge/full CI is needed tpu Related to Google TPUs v1
#21128 opened Jul 17, 2025 by DarkLight1337 Loading…
2 of 4 tasks
17
docker: docker-aware precompiled wheel support ci/build
#21127 opened Jul 17, 2025 by dougbtv Loading…
4 tasks done
8
[WIP] Use FlashInfer RoPE
#21126 opened Jul 17, 2025 by mgoin Loading…
4 tasks
3
[Refactor] Remove Unused Naive Moe Kernels performance Performance-related issues
#21125 opened Jul 17, 2025 by yewentao256 Loading…
1
2
[UPDATED] - Large Block_size solution v1
#21123 opened Jul 17, 2025 by nadathurv Loading…
3
[Bugfix] Allocate less memory in non-batched CUTLASS MoE bug Something isn't working ready ONLY add when PR is ready to merge/full CI is needed
#21121 opened Jul 17, 2025 by ElizaWszola Loading…
5
security policy: take 1 documentation Improvements or additions to documentation
#21119 opened Jul 17, 2025 by sidhpurwala-huzaifa Loading…
7
feat: add fused MLA QKV + strided layernorm deepseek Related to DeepSeek models
#21116 opened Jul 17, 2025 by mickaelseznec Loading…
3 of 4 tasks
10
ProTip! Filter pull requests by the default branch with base:main.