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
- [Refactor] RoPE: Support RotaryEmbedding class usage for DiT models
- Quick survey: your experience as a PyTorch Ecosystem maintainer
- [Fix] wan2.2-s2v: Take the last wav2vec2 layer from last_hidden_state
- [Perf] wan2.2-s2v: Compute RoPE in fp32 via _apply_rotary_emb, as in wanvideo.py
- [Perf] glm47: Match streaming tool calls with str.find instead of a lazy regex
- [Perf] pythonic: Resume streaming bracket matching instead of rescanning the buffer
- Draft 1016
- [Feat] dwdp: Support MTP speculative decoding on DWDP prefill
- [Fix] openai: Enforce --trust-request-chat-template on responses, decisions and systemone
- [Fix] scheduler: Allow memory release with paused grammar compilation
- Docs
- Python not yet supported