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
- [diffusion] Expose MiniMax-H3 AdaLN cache stats in perf dumps
- [Fix] Handle whitespace in Qwen3 Coder boolean tool arguments
- Avoid server model imports in the serving benchmark client
- [DSpark] Select DSv4.1 candidate verify graphs from committed lengths under compact verify
- [DCP] Support block-FP8 Q projection replication
- [Perf] Enable fast-math for JIT activations on Blackwell
- [HiCache] NVFP4 KV + kernel io-backend: host pool row geometry mismatch and block scales never transported (packed MTP draft silently corrupted, accept length pinned to 1.0)
- fix(hicache): square NVFP4 host-pool geometry + page-granular block-scale transport
- [DeepSeek-V4.1] Serve image requests under encoder SWA bounded replay
- [diffusion] keep TP/Ulysses head shards on the unsharded torch SDPA kernel
- Docs
- Python not yet supported