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
- [DLLM] DLLM_EXTEND through FlashInfer block_extend attention
- [RFC] Pluggable KV Compression for Disaggregated Serving
- [Fix] Fallback to unquantized linear method for Qwen3.5 GDN in_proj_ba (#38854)
- Fix PD retraction checkpoint boundaries with overlap decoding
- fix(hicache): back up hybrid-Mamba prefixes whose last node ends exactly on a chunk boundary
- feat(hicache): make the storage existence-cache capacity configurable (SGLANG_HICACHE_EXISTENCE_CACHE_MAX_ENTRIES)
- Preserve full DFlash draft KV during PD CPU retraction
- [Feature] Forward raw token completions and expose backend model lengths
- Fix fallback signatures for Anthropic thinking blocks
- [Bugfix] Make lazy kernel resolution compatible with torch.compile fullgraph
- Docs
- Python not yet supported