Skip to content

Pull requests: vllm-project/vllm

New pull request New
605 Open 8,003 Closed
605 Open 8,003 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[Misc] add get kv cache token capacity frontend v1
#17538 opened May 1, 2025 by lengrongfu • Changes requested
1
1
[Model] Refactor Ovis2 to support original tokenizer documentation Improvements or additions to documentation ready ONLY add when PR is ready to merge/full CI is needed
#17537 opened May 1, 2025 by Isotr0py • Review required
2
[CI/Build] Remove awscli dependency ci/build ready ONLY add when PR is ready to merge/full CI is needed
#17532 opened May 1, 2025 by DarkLight1337 • Review required
3
Move the last arguments in arg_utils.py to be in their final groups ready ONLY add when PR is ready to merge/full CI is needed
#17531 opened May 1, 2025 by hmellor Loading…
4
[FEAT][ROCm]: Support AITER MLA on V1 Engine ci/build rocm Related to AMD ROCm v1
#17523 opened May 1, 2025 by vllmellm Loading…
1
[prototype] prioritized block soft pinning/evictions documentation Improvements or additions to documentation frontend v1
#17520 opened May 1, 2025 by simon-mo • Draft
1
[V1] Add num_cached_tokens stats for request output ready ONLY add when PR is ready to merge/full CI is needed v1
#17519 opened May 1, 2025 by simon-mo Loading…
2
[Bugfix][Model] vllm-v0 engine run eagle algo with qwen2.5 model, KeyError: 'norm.weight' bugfix
#17518 opened May 1, 2025 by Greatpanc Loading…
1
1
[Bugfix][V1][Spec Dec] Add generator to request even when no seed is provided. speculative-decoding v1
#17509 opened May 1, 2025 by luyuzhe111 Loading…
4
[Frontend][TPU] Add TPU default max-num-batched-tokens based on device name
#17508 opened May 1, 2025 by Chenyaaang Loading…
5
[BugFix] Qwen3 tool calling failed using qwen3 reasoning parser. documentation Improvements or additions to documentation frontend tool-calling
#17506 opened Apr 30, 2025 by Xu-Wenqing Loading…
4
[V1][Spec Decode] Apply torch.compile & cudagraph to EAGLE3 ready ONLY add when PR is ready to merge/full CI is needed v1
#17504 opened Apr 30, 2025 by zixi-qi Loading…
10
[RFC][core][V1] generalize structured output manager and backends structured-output tpu Related to Google TPUs v1
#17503 opened Apr 30, 2025 by william-baker-inflection Loading…
1
[Chore] Ignore Ruff warning on E501
#17502 opened Apr 30, 2025 by aarnphm Loading…
11
[Model] Add GraniteMoeHybrid 4.0 model
#17497 opened Apr 30, 2025 by s3woz Loading…
10
[TPU] Add kernel test for moe_pallas ci/build ready ONLY add when PR is ready to merge/full CI is needed tpu Related to Google TPUs
#17496 opened Apr 30, 2025 by mgoin Loading…
1
[Bugfix] Adding maxnreg to lora expand/shrink kernel definition
#17492 opened Apr 30, 2025 by ozziemoreno Loading…
1
1
Fix arg checking for GGUF/Quark/GPTQMarlin quantized MoE methods bug Something isn't working ready ONLY add when PR is ready to merge/full CI is needed
#17491 opened Apr 30, 2025 by mgoin Loading…
1
[V1] Add VLLM_ALLOW_INSECURE_SERIALIZATION env var v1
#17490 opened Apr 30, 2025 by russellb Loading…
1
Add full API docs and improve the UX of navigating them ci/build documentation Improvements or additions to documentation frontend multi-modality Related to multi-modality (#4194) v1
#17485 opened Apr 30, 2025 by hmellor Loading…
6
[Attention] MLA move o_proj q_proj into cuda-graph region v1
#17484 opened Apr 30, 2025 by LucasWilkinson Loading…
2
[v1] Pass BlockTable and KVCacheSpec to AttentionMetadataBuilders tpu Related to Google TPUs v1
#17483 opened Apr 30, 2025 by heheda12345 Loading…
2
[v1] Introduce KVCacheBlocks as interface between Scheduler and KVCacheManager v1
#17479 opened Apr 30, 2025 by heheda12345 Loading…
2
[v1] Move block management logic from KVCacheManager to SpecializedManager v1
#17474 opened Apr 30, 2025 by heheda12345 Loading…
2
fix: restore http metrics for V0 engine v1
#17471 opened Apr 30, 2025 by davidxia • Draft
1
8
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.