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
- [Bug] Tokenizer workers open CUDA contexts on the base GPU despite --mm-feature-transport cpu
- Keep tokenizer workers off the base GPU under CPU feature transport
- [AMD] Extend DSV4 ROCm multi-stream overlap to HCA
- [1/N] Introduce a Fault-Tolerance Framework for SGLang
- [Rust] Fix listener port collisions for system-DP replicas
- [Bug] developer messages are silently dropped by chat templates without role support
- [sgl-router] Policy reorg 3/N: bucket-engine core (Policy::pick, admission, load-only policies)
- [sgl-router] Policy reorg 4/N: affinity store, session-aware and sticky policies
- [sgl-router] Policy reorg 5/N: cache-aware policy with a local/remote prefix source
- [sgl-router] Policy reorg 6/N: per-bucket policy in config and --selection-engine
- Docs
- Python not yet supported