Skip to content

Pull requests: vllm-project/vllm

New pull request New
2,229 Open 21,373 Closed
2,229 Open 21,373 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

[Renderer] Remove InputPreprocessor ready ONLY add when PR is ready to merge/full CI is needed v1
#38688 opened Apr 1, 2026 by DarkLight1337 Loading…
5 tasks
5
fix(lora): use float32 intermediate buffer in fused MoE LoRA to prevent bf16 precision loss
#38686 opened Apr 1, 2026 by prsabahrami • Draft
2
[ROCm][CI] Remove soft_fail from AMD Docker Image Build ci/build rocm Related to AMD ROCm
#38685 opened Apr 1, 2026 by micah-wil • Draft
[Perf] DSV3.2 Indexer Fused Weights Projection deepseek Related to DeepSeek models
#38684 opened Apr 1, 2026 by benchislett Loading…
5
[Quantization] Rename mxfp4 quant layer and oracle to gpt_oss_mxfp4 gpt-oss Related to GPT-OSS models
#38683 opened Apr 1, 2026 by zyongye Loading…
3 of 5 tasks
[XPU] [Quant] rename mxfp8_e4m3_quantize and add xpu backend implementation intel-gpu Related to Intel GPU
#38682 opened Apr 1, 2026 by zufangzhu Loading…
6
[CPU] Fix lscpu NUMA node regex to handle quoted - and null in containers cpu Related to CPU backends
#38681 opened Apr 1, 2026 by Monokaix Loading…
3 of 5 tasks
[CI][ROCm] Remove unsupported cases in test_fusion.py rocm Related to AMD ROCm
#38680 opened Apr 1, 2026 by charlifu Loading…
fused_moe_kernel opt
#38679 opened Apr 1, 2026 by SYChen123 Loading…
5 tasks
3
Fix llm_request trace context propagation frontend v1
#38678 opened Apr 1, 2026 by will-deines Loading…
[CPU] Support head_size 512 in cpu_attn cpu Related to CPU backends documentation Improvements or additions to documentation ready ONLY add when PR is ready to merge/full CI is needed v1
#38676 opened Apr 1, 2026 by bigPYJ1151 Loading…
5 tasks
2
[torch.compile] Add compile-only mode frontend v1
#38675 opened Apr 1, 2026 by zou3519 • Draft
10
[Bugfix] Preserve original ImportError in gRPC server entrypoint bug Something isn't working frontend
#38673 opened Apr 1, 2026 by CatherineSue Loading…
3 tasks done
[5/n] Migrate CUTLASS MLA, hadamard, awq, allspark and DSV3 fused a gemm to torch stable ABI ci/build nvidia
#38671 opened Apr 1, 2026 by mikaylagawarecki • Draft
5 tasks
8
[Bugfix] Fix AWQ models batch invariance issues bug Something isn't working v1
#38670 opened Apr 1, 2026 by YM2132 Loading…
5 tasks done
1
4
Fix Marlin repack PTX incompatibility on H100/H200 (CUDA 12.8) ci/build nvidia
#38669 opened Apr 1, 2026 by DavidBellamy Loading…
1
4
add vLLM-side LMCache EC connector entrypoint kv-connector
#38668 opened Mar 31, 2026 by benyebai • Draft
4
[ROCm] Enable dual-stream MoE shared experts and GLM-5 MXFP4 Quark support rocm Related to AMD ROCm v1
#38665 opened Mar 31, 2026 by ChuanLi1101 Loading…
4 tasks
3
[CI][ROCm] Add Qwen3.5-35B-A3B-MXFP4 model eval into CI qwen Related to Qwen models rocm Related to AMD ROCm
#38664 opened Mar 31, 2026 by BowenBao • Draft
4
[Core][Feat][ safely abort requests where FSM failed to advance v1
#38663 opened Mar 31, 2026 by walterbm Loading…
3 of 5 tasks
2
[Kernel] feat: TurboQuant KV cache quantization (PolarQuant + QJL) ci/build v1
#38662 opened Mar 31, 2026 by allaspectsdev Loading…
1 of 6 tasks
6
[2/N] Pass model_config to the Attention constructors deepseek Related to DeepSeek models gpt-oss Related to GPT-OSS models llama Related to Llama models qwen Related to Qwen models ready ONLY add when PR is ready to merge/full CI is needed
#38661 opened Mar 31, 2026 by MatthewBonanni Loading…
3 of 5 tasks
1
[compile] Invoke split FX graph by codegen.
#38657 opened Mar 31, 2026 by zhxchen17 Loading…
5 tasks
6
Fix Nano Nemotron VL regressions multi-modality Related to multi-modality (#4194)
#38655 opened Mar 31, 2026 by netanel-haber Loading…
8
[Bugfix] Fix vllm bench serve to count multimodal tokens in "total input tokens" bug Something isn't working performance Performance-related issues
#38654 opened Mar 31, 2026 by mgehre-amd Loading…
6
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.