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
- [Fix] flashinfer-megamoe: Fix startup autotune on FlashInfer 0.7.2 and add an optional decode workspace
- [Fix] flashinfer: Apply the sliding window in ragged prefill for hybrid-SWA models
- [Fix] hunyuan: Escape streamed string arguments like json.dumps
- [Fix] qwen3_coder: Close streamed arguments when the model omits </function>
- [Fix] scheduler: Keep the paged intake limit admittable on an empty KV pool
- [Fix] attention: Zero padded KDA/GDN recurrent outputs
- [Fix] sgl-router: Run the chat render parity generator through deepseek-recipe
- [Perf] sgl-router: Encode tiktoken models with fastokens under --tokenizer-backend fast
- [Fix] diffusion: LoRA alpha reload, auto-mode skip, critic coverage
- [Perf] deepseek-v4: Reuse combined sparse-prefill indices
- Docs
- Python not yet supported