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
- [Feat] disagg: Admit the DeepSeek-V4 paged KV layout in the PD decode radix cache
- [Security] Fix multimodal SSRF local file disclosure with path allowlist and validation (#45206)
- [elastic ep] live serving during scale
- [Feat] external linker: support lmcache external linker
- [Bug] HiCache NIXL L3: under O_DIRECT the DSA indexer sidecar is never written, and the 0-byte objects are restored as hits, so restored prefixes get zero index-K (silent output change, EAGLE accept collapse)
- [RFC] Attention-FFN disaggregation: design and integration plan
- [Feature] Support symmetric static compressed-tensors W8A8 on Ascend
- [Feat] scheduler: Persist hybrid active requests for fresh-process resume
- [Perf] [AMD] wip_rccl: upgrade rccl to 2.32.3
- [Fix] scheduler: Fence a chunk stash that frees duplicate KV behind the in-flight forward
- Docs
- Python not yet supported