sglang
https://github.com/sgl-project/sglang
Python
SGLang is a high-performance serving framework for large language models and multimodal models.
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 supported1 Subscribers
Add a CodeTriage badge to sglang
Help out
- Issues
- [Feature][Spec][XPU] Honor rejection sampling in EAGLE chain verification
- [AMD] Route gfx942 block-FP8 linears to the AITER CK blockscale GEMM
- [Bugfix] Skip post-experts EP all-reduce on the FlashInfer cutlass FP4 all-gather MoE path
- [Diffusion] Add native MLX support for Qwen-Image 2.1
- [Bug] DWDP: expert weight copy peaks at 2x local expert memory, causing OOM at startup
- [OpenAI] Forward request priority through /v1/responses
- fix(rotary_embedding): support partial rotary embeddings in forward_xpu fallback
- [Bug] /v1/responses drops the request priority
- [DSV4.1] Add H20 block-FP8 tuned configs
- fix(bench): speed-bench counts 2 input tokens per prompt when the chat template returns BatchEncoding
- Docs
- Python not yet supported