Skip to content

Pull requests: vllm-project/vllm

New pull request New
1,254 Open 14,570 Closed
1,254 Open 14,570 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] Enable Expert parallel for MoE models
#28263 opened Nov 7, 2025 by jikunshang Loading…
5 tasks
3
[CPU]Avoid repeated random sample compile v1
#28260 opened Nov 7, 2025 by xiangze-arm Loading…
3
[Misc][Model][Refactor] Pass the prefix into Linear layers deepseek Related to DeepSeek models qwen Related to Qwen models ready ONLY add when PR is ready to merge/full CI is needed
#28259 opened Nov 7, 2025 by MengqingCao Loading…
5 tasks
6
[GPUModelRunner] initialize_kv_cache cleanup (1/N): move initialization that doesn't depend on kv cache config to load_model v1
#28258 opened Nov 7, 2025 by heheda12345 Loading…
5 tasks
4
Fix issues from #28242 qwen Related to Qwen models ready ONLY add when PR is ready to merge/full CI is needed
#28257 opened Nov 7, 2025 by hmellor Loading…
3
[Bug] Fix missing token_ids for reasoning parser models in chat completions #28246 frontend
#28256 opened Nov 7, 2025 by baonudesifeizhai Loading…
5 tasks
3
[Log] update shm wait time msg ready ONLY add when PR is ready to merge/full CI is needed
#28255 opened Nov 6, 2025 by BoyuanFeng Loading…
5
[BugFix] Fix DeepGEMM over-allocating workspace
#28254 opened Nov 6, 2025 by LucasWilkinson Loading…
1
[BugFix] Avoid calling KV connector layer APIs when metadata is unset kv-connector
#28253 opened Nov 6, 2025 by sdavidbd Loading…
1
1
[WIP][KVConnector] Retrive KV events from LMCache kv-connector
#28252 opened Nov 6, 2025 by hickeyma • Draft
1 of 5 tasks
6
[CI/Build] add src folder to AMD test Dockerfile to fix python_only_compile ci/build rocm Related to AMD ROCm
#28251 opened Nov 6, 2025 by bradleyhd Loading…
5 tasks
4
[Core] Rework handling of async scheduling config ready ONLY add when PR is ready to merge/full CI is needed v1
#28250 opened Nov 6, 2025 by njhill Loading…
8
parse reasoning item input in responses api frontend gpt-oss Related to GPT-OSS models
#28248 opened Nov 6, 2025 by qandrew • Draft
[Perf] Use np.ndarray instead of list[list[int]] to reduce GC overhead ready ONLY add when PR is ready to merge/full CI is needed v1
#28245 opened Nov 6, 2025 by Jialin Loading…
3 of 5 tasks
4
Add truncate arg to yarn to match openai implementation of gpt-oss gpt-oss Related to GPT-OSS models
#28244 opened Nov 6, 2025 by ashors1 Loading…
5 tasks
10
Fix CPU offload KV cache accounting and align CUDA xformers pin ci/build v1
#28241 opened Nov 6, 2025 by m0nk111 Loading…
1
4
[BugFix][27485] Fix ITL algorithm for chunked OpenAI chat completions performance Performance-related issues
#28240 opened Nov 6, 2025 by manamalani10 Loading…
1
3
[NVIDIA] [feat] Integrate flashinfer Trtllmgen bf16 moe ci/build documentation Improvements or additions to documentation qwen Related to Qwen models
#28238 opened Nov 6, 2025 by jiahanc • Draft
5 tasks
1
[Frontend][responsesAPI] convert responses API tool input to chat completions tool format frontend
#28231 opened Nov 6, 2025 by qandrew Loading…
4
[Feature] Default ignore_eos True for random dataset performance Performance-related issues ready ONLY add when PR is ready to merge/full CI is needed
#28227 opened Nov 6, 2025 by yewentao256 Loading…
5
[BugFix] [FEAT] Enable fastsafetensors for ROCm platform ci/build rocm Related to AMD ROCm
#28225 opened Nov 6, 2025 by tjtanaa Loading…
5 tasks
6
Update xgrammar version from 0.1.25 to 0.1.27 ci/build
#28221 opened Nov 6, 2025 by cjackal Loading…
1 task done
1
Adds Dockerfile arg for VLLM_PRECOMPILED_WHEEL_LOCATION ci/build
#28217 opened Nov 6, 2025 by dougbtv Loading…
3 tasks done
3
Fix cu_num_generated_tokens slicing logic in LogprobsLists.slice() method v1
#28214 opened Nov 6, 2025 by usberkeley Loading…
5 tasks done
2
Enhance Helm chart installation instructions documentation Improvements or additions to documentation
#28211 opened Nov 6, 2025 by ccnmxns Loading…
7
ProTip! Exclude everything labeled bug with -label:bug.