vllm
https://github.com/vllm-project/vllm
Python
A high-throughput and memory-efficient inference and serving engine for LLMs
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported25 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- Support TurboQuant for YOCO + sliding-window models (e.g., Gemma 4 E4B)
- [MoE] [MXFP4] refactor compressed tensor w4a4 mxfp4 moe
- [ROCm] torch 2.11 + rocclr profiler hotfix for ROCm 7.2 + aiter v0.1.12.post2
- fix(docker): add SM103a (GB300) to DeepEP CUDA arch list
- Integrate flashinfer b12x MoE and FP4 GEMM kernels for SM120/121
- [Attention] Add SM120 attention sink support via flashinfer fmha_v2
- Added vLLM support for BitNet b1.58 2B4T
- [Model] support Qianfan-OCR model
- [Bugfix][CI] Fix `from . import nixl_ep_cpp as _nixl_ep_cpp`
- Quantized fused MOE optimization
- Docs
- Python not yet supported