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
- [Feature] Forward unknown tool calls by default; the parse-time drop kills agent loops
- [Perf] Extend Kimi K3 GEMM/all-gather to the speculative verification front
- [Tool Call] Forward unknown tools to clients by default
- fix(models): guard both format_ue8m0 reads in deepseek weight loader (fixes humming weight-load crash)
- test: cover DSV4 history tool-call argument encoding
- [Bugfix] Harden DeepSeek V4 reasoning and DSML tool-call parsing
- docs: refresh README news and community links
- [Bug] LoRA adapters with use_rslora=True are served with the wrong scale
- [LoRA] Scale rsLoRA adapters by alpha / sqrt(r)
- [quant] fix: accept a pre-quantized input in FlashInfer's block-FP8 GEMM
- Docs
- Python not yet supported