Skip to content

Pull requests: vllm-project/vllm

New pull request New
1,984 Open 20,397 Closed
1,984 Open 20,397 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

[CI] Split Split Distributed Tests (4 GPUs), Entrypoints and Kernel MoE tests ci/build ready ONLY add when PR is ready to merge/full CI is needed
#37100 opened Mar 15, 2026 by avinashsingh77 Loading…
3 tasks
4
Unwrap fused_moe (control via env var) for non-DP-EP cases
#37099 opened Mar 15, 2026 by SouthWest7 • Draft
1 of 5 tasks
7
Fix Hermes streaming for parameterless tool args tool-calling
#37098 opened Mar 15, 2026 by Shaunak00 Loading…
1
Use contiguous arrays for request token histories v1
#37097 opened Mar 15, 2026 by Shaunak00 Loading…
3
[Profiling] Add optional stage-level NVTX annotations for Nsight Systems documentation Improvements or additions to documentation v1
#37095 opened Mar 15, 2026 by KOKOSde • Draft
5
[Bugfix] Align routed_experts with streamed and final output token spans bug Something isn't working v1
#37094 opened Mar 15, 2026 by HareshKarnan Loading…
7
[Frontend][Misc] Remove unused log in /is_sleeping frontend
#37093 opened Mar 15, 2026 by esmeetu Loading…
5 tasks
2
[WIP][Bugfix] Clamp -1 async placeholders to fix CUDA assert in multimodal+EAGLE3 bug Something isn't working nvidia v1
#37092 opened Mar 15, 2026 by haosdent • Draft
1
[WIP][Bugfix] Add autotuning guard to all unprotected FlashInfer MoE kernels bug Something isn't working nvidia
#37091 opened Mar 15, 2026 by haosdent • Draft
2 of 3 tasks
1
7
[WIP][Bugfix] Disable cross-layer KV cache for MLA attention backends bug Something isn't working v1
#37090 opened Mar 15, 2026 by haosdent • Draft
1
[WIP] [Model] Support Qwen1 use_logn_attn and use_dynamic_ntk qwen Related to Qwen models
#37089 opened Mar 15, 2026 by haosdent • Draft
1
2
[Bugfix] Move GDN warmup after KV cache allocation to fix memory leak (#36973) bug Something isn't working qwen Related to Qwen models
#37088 opened Mar 15, 2026 by haosdent Loading…
6
feat: Standardize load_weights API via AutoWeightsLoader documentation Improvements or additions to documentation llama Related to Llama models
#37085 opened Mar 15, 2026 by akh64bit Loading…
1
18
[Misc] Add unit tests for min_p Triton sampling kernel v1
#37083 opened Mar 15, 2026 by gkuwanto Loading…
3 of 5 tasks
4
[V1, V2] Add temperature for prompt logprobs frontend v1
#37082 opened Mar 15, 2026 by JacobHelwig Loading…
3 of 5 tasks
5
Add Mistral Guidance frontend structured-output v1
#37081 opened Mar 14, 2026 by juliendenize Loading…
3 tasks done
5
[Feature][Cleanup]: Optimize token data structures (list[int] to TokenArray) kv-connector structured-output tpu Related to Google TPUs v1
#37078 opened Mar 14, 2026 by akh64bit Loading…
1
18
[Feature][Frontend] add support for Cohere Embed v2 API documentation Improvements or additions to documentation frontend
#37074 opened Mar 14, 2026 by walterbm Loading…
4 of 5 tasks
6
[Bugfix] Fix harmony parser crash on terminal tokens after end-of-message bug Something isn't working frontend gpt-oss Related to GPT-OSS models
#37072 opened Mar 14, 2026 by Pradyun92 Loading…
3 of 5 tasks
4
[Bugfix] Fix Responses API harmony streaming: token splitting, missing done events, nested sequence_number bug Something isn't working frontend gpt-oss Related to GPT-OSS models
#37071 opened Mar 14, 2026 by Pradyun92 Loading…
3 of 5 tasks
2
[Bugfix] Fix harmony streaming tool call crash and argument splitting bug Something isn't working frontend gpt-oss Related to GPT-OSS models
#37070 opened Mar 14, 2026 by Pradyun92 Loading…
3 of 5 tasks
Yzong rh/nixl ssm rebase kv-connector v1
#37069 opened Mar 14, 2026 by yzong-rh • Draft
3
[Bugfix] Add Qwen3.5 MoE support to benchmark_moe.py bug Something isn't working performance Performance-related issues qwen Related to Qwen models
#37068 opened Mar 14, 2026 by iphands Loading…
4
Improve CPU platform detection fallback for source checkouts
#37067 opened Mar 14, 2026 by ezylopx5 Loading…
3
[Distributed] Add OfflineState bloom-filter cooperative caching KV connector documentation Improvements or additions to documentation kv-connector performance Performance-related issues
#37066 opened Mar 14, 2026 by Ilank1 Loading…
4 tasks
6
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.