Skip to content

Pull requests: vllm-project/vllm

New pull request New
2,517 Open 22,209 Closed
2,517 Open 22,209 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

[Docker] Non-root support for vllm-openai; add opt-in vllm-openai-nonroot target ci/build documentation Improvements or additions to documentation
#40275 opened Apr 19, 2026 by TheDuyIT Loading…
3 of 4 tasks
1
4
[Doc] Fix broken "full example" links in structured_outputs.md documentation Improvements or additions to documentation structured-output
#40274 opened Apr 19, 2026 by FrimpsManu Loading…
1
3
Fix MoE backend selection for LoRA (unquantized MoE)
#40273 opened Apr 19, 2026 by danisereb Contributor • Draft
4 tasks
2
docs: expand load_weights guide with AutoWeightsLoader and manual patterns documentation Improvements or additions to documentation
#40271 opened Apr 19, 2026 by RudrenduPaul Loading…
3 tasks
1
8
【Feat】GPU KV Cache Pluggable Eviction Policy v1
#40270 opened Apr 19, 2026 by sjmshsh Loading…
4
[Bugfix][Spec Decode] Wire draft_probs into probabilistic draft_model rejection bug Something isn't working speculative-decoding v1
#40269 opened Apr 19, 2026 by bedeks Loading…
4 tasks
1
1
refactor: make is_neox_style configurable for glm deepseek Related to DeepSeek models
#40267 opened Apr 19, 2026 by inisis Loading…
4 tasks
1
[Doc] Fix typos in token_embed pooling documentation documentation Improvements or additions to documentation
#40266 opened Apr 19, 2026 by YifanLi3 Loading…
2
Fix tp device index overflow ci/build cpu Related to CPU backends documentation Improvements or additions to documentation frontend kv-connector llama Related to Llama models multi-modality Related to multi-modality (#4194) needs-rebase new-model Requests to new models nvidia qwen Related to Qwen models speculative-decoding tool-calling v1
#40265 opened Apr 19, 2026 by kari-smasint Loading…
2 tasks
15
[ROCm] Profiler api support for ROCm MORI toy proxy server in PD Disaggregation documentation Improvements or additions to documentation kv-connector rocm Related to AMD ROCm
#40264 opened Apr 19, 2026 by itej89 Contributor Loading…
3 of 4 tasks
7
[Doc] Fix CLI help examples: remove phantom --help=listgroup and --help=page modes documentation Improvements or additions to documentation
#40262 opened Apr 18, 2026 by avasis-ai Loading…
2 tasks done
2
[Bugfix] Corrects estimate of torch memory use causing OOM due to incorrect KV cache space estimation when sleep mode on (Fixes #40256) bug Something isn't working
#40258 opened Apr 18, 2026 by djparente Loading…
3 of 4 tasks
5
Fix view shape assignment for chan_scales
#40257 opened Apr 18, 2026 by zjysteven Loading…
3 of 4 tasks
4
fix(libtorch_stable): guard mxfp4 ops behind ENABLE_NVFP4_SM100
#40255 opened Apr 18, 2026 by svilendotorg Loading…
1
[ROCm] Add missing gfx1152, gfx1153, and enable all gpu arch to AITER in docker ci/build documentation Improvements or additions to documentation rocm Related to AMD ROCm
#40254 opened Apr 18, 2026 by thelittlefireman Loading…
9
[Bugfix] Stream MiniMax M2 tool-call deltas incrementally bug Something isn't working frontend tool-calling
#40253 opened Apr 18, 2026 by wuyingjun-lucky Loading…
1
6
[Bugfix] Forward mm_processor_kwargs in offline generate APIs bug Something isn't working frontend
#40251 opened Apr 18, 2026 by wuyingjun-lucky Loading…
1
4
fix: remove redundant None default in dict.get() calls documentation Improvements or additions to documentation kv-connector
#40250 opened Apr 18, 2026 by hummbl-dev Loading…
2
[torch.compile] refactor config hashing through compile_factors and normalization llama Related to Llama models v1
#40246 opened Apr 18, 2026 by WorldExplored Contributor Loading…
3
[Qwen][Bugfix] Fixes sigmoid activation in torch impl of RMSNormGated. bug Something isn't working qwen Related to Qwen models ready ONLY add when PR is ready to merge/full CI is needed
#40245 opened Apr 18, 2026 by sighingnow Collaborator Loading…
3
[Bugfix] GLM tool parser: fix streaming corruption for Optional[str]/array args bug Something isn't working tool-calling
#40197 opened Apr 18, 2026 by kulpsin Loading…
3 of 4 tasks
1
3
[Bugfix] Make Attention Backend Auto-Selection Batch-Invariance-Aware bug Something isn't working v1
#40193 opened Apr 18, 2026 by WorldExplored Contributor Loading…
1
4
[Frontend] Add defer_loading and tool_reference support for Anthropic and OpenAI APIs frontend ready ONLY add when PR is ready to merge/full CI is needed
#40190 opened Apr 18, 2026 by JaredforReal Contributor Loading…
4 tasks
12
[Distributed] Add MSCCL++ allreduce support for multi-node communication nvidia v1
#40188 opened Apr 18, 2026 by AO114 Loading…
4
[torch.compile] Remove layer name from unified_kv_cache_update / unified_mla_kv_cache_update to fix cold-start (#33267)
#40187 opened Apr 18, 2026 by Doondi-Ashlesh Loading…
2 of 4 tasks
1
2
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.