Skip to content

Pull requests: vllm-project/vllm

New pull request New
1,485 Open 18,183 Closed
1,485 Open 18,183 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

[Nightly CI] Remove CT Model
#33530 opened Feb 2, 2026 by robertgshaw2-redhat Loading…
1
Triton MLA GQA perf fixes (4x improvement at 80k context) v1
#33529 opened Feb 2, 2026 by koush Loading…
3 of 5 tasks
11
Adds padding and perf improvements to wvSplitK_fp8 rocm Related to AMD ROCm
#33527 opened Feb 2, 2026 by amd-hhashemi Loading…
5 tasks
6
Update get_expert_mapping to include self parameter
#33525 opened Feb 1, 2026 by Otsutsukii Loading…
5 tasks
1
3
[Core][Scheduler] Fix FCFS queue ordering for skipped waiting requests v1
#33522 opened Feb 1, 2026 by harsh543 Loading…
1
2
Fix mistral sliding window parsing
#33521 opened Feb 1, 2026 by andylolu2 Loading…
7
[Bugfix] Fix tool call streaming for gpt-oss/Harmony models bug Something isn't working frontend gpt-oss Related to GPT-OSS models
#33520 opened Feb 1, 2026 by alexbi29 Loading…
@chaunceyjiang
6
[Bugfix] Fix NVFP4 MoE weight shapes for non-gated MLPs (Nemotron-Nano) bug Something isn't working nvidia
#33518 opened Feb 1, 2026 by Code4me2 Loading…
1
3
[Kernel] Add enable_sm120_or_later for SM121 (DGX Spark) CUTLASS support nvidia
#33517 opened Feb 1, 2026 by Code4me2 Loading…
1
4
[Bugfix] Add SM110/SM120 device capability checks for NVFP4 MoE backends bug Something isn't working nvidia
#33516 opened Feb 1, 2026 by Code4me2 Loading…
3
7
[Bugfix] Fix gpt-oss chat format mismatch with HuggingFace bug Something isn't working frontend gpt-oss Related to GPT-OSS models
#33514 opened Feb 1, 2026 by thjung123 Loading…
5 tasks
5
Fix reasoning_tokens for text-based parsers in Responses API frontend
#33513 opened Feb 1, 2026 by anencore94 Loading…
1
4
fix(ROCm): Make flash_attn import optional in MLA attention rocm Related to AMD ROCm
#33511 opened Feb 1, 2026 by rabi Loading…
6
[FIX] guidance: use max(vocab_size, len(tokenizer)) for n_vocab structured-output v1
#33509 opened Feb 1, 2026 by FredericOdermatt Loading…
1 of 5 tasks
5
Support FI fused MoE non gated FP8 & NVFP4 nvidia
#33506 opened Feb 1, 2026 by amitz-nv • Draft
5 tasks
4
[Feature]: Qwen3-Next dual-stream execution in_proj_qkvz in_proj_ba qwen Related to Qwen models
#33505 opened Feb 1, 2026 by SouthWest7 • Draft
1 of 5 tasks
2
Add **kwargs parameter to v1 FlashAttentionImpl as catch-all v1
#33504 opened Feb 1, 2026 by haojin2 Loading…
5 tasks done
6
feat(spec_decode): fuse EAGLE step slot mapping and metadata updates speculative-decoding v1
#33503 opened Feb 1, 2026 by sladyn98 Loading…
3 of 5 tasks
6
Scheduler: skip KV-blocked requests to prevent blocking (#31731) v1
#33499 opened Feb 1, 2026 by harsh543 Loading…
1
3
[Experimental][Refactor] Refactor vision chunk modality processing for unification documentation Improvements or additions to documentation multi-modality Related to multi-modality (#4194) needs-rebase
#33498 opened Feb 1, 2026 by Isotr0py • Draft
5 tasks
7
[Bugfix] Fix assertion error in flashmla backend with fullgraph enabled bug Something isn't working v1
#33496 opened Feb 1, 2026 by Kurumi5210 Loading…
5 tasks
1
[Feature] Support 'int8' in CacheConfig validation
#33495 opened Feb 1, 2026 by drshvik Loading…
6
[Doc]: update paths for Offline/Online/Others example sections documentation Improvements or additions to documentation
#33494 opened Feb 1, 2026 by soyr-redhat Loading…
4 of 5 tasks
4
Perf tuning and expansion of cases covered for wvSplitKrc rocm Related to AMD ROCm
#33493 opened Feb 1, 2026 by amd-hhashemi Loading…
5 tasks
7
Sort safetensors files to ensure deterministic loading order
#33491 opened Feb 1, 2026 by Lumosis Loading…
1 of 5 tasks
3
ProTip! Filter pull requests by the default branch with base:main.