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] preserve embedding priority in batch slicing
- [NPU] Fix double Mamba ping-pong flip after spec verify
- [Model] Add MageVL multimodal model support
- [Perf] Capture target-verify metadata rebuild inside the decode cuda graph
- [HiCache] NIXL storage registration acquire/release (RFC #32841, stage 1)
- fix(distributed): use cpu group for custom parallel config exchange
- [NCCL RAS] Integrate runtime diagnostics into health checks (detection + reporting)
- fix(dspark): honor partial budgets in full-window fast path
- [AMD] perf: fuse Kimi-K2.5 MLA projection, RoPE, and KV-cache write
- fix(dsv4): flatten OpenAI list content to string in encoder
- Docs
- Python not yet supported