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
- [HiCache] Buffer-only L3 for an auxiliary SWA window; load back only missing window rows
- [dLLM] Add a runtime structure walkthrough (stack 4/4)
- Accept whole-number floats for integer tool arguments in the response parser
- [Rust] Serve scheduler metrics on the Rust frontend's /metrics
- [sglang-miles] Return routed experts as a compressed, self-describing payload
- [Fix] hicache: Bound staged admission retries on shared capacity rejection
- [AMD] [Fix] Prevent ComponentResidencyError after compile warmup with explicit DiT layerwise offload
- [Fix] [AMD] dit: Apply model-default prefetch to explicit layerwise offload
- [Bugfix] Honor SGLANG_NGRAM_FORCE_GREEDY_VERIFY for NGRAM
- [NPU] Fix DeepSeek-V4 hc_pre in dual-machine PD disaggregation on Ascend 950
- Docs
- Python not yet supported