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 Glm4MoeDetector streaming emitting invalid JSON for non-string args
- Foundry Adapter
- [Test] diffusion: Cover residual dispatch and Qwen CUDA fallbacks
- [Fix] diffusion/lora: Preserve eager base dispatch for disabled and merged adapters
- [Bug] Anthropic streaming drops a tool call's closing "}" when whitespace text arrives between its arguments (multiple tool calls, qwen3_coder)
- [Fix] Parse coalesced DeepSeek V3.1 streaming tool calls
- [sgl-kernel] Build common_ops for sm_90a when FA3 is off (aarch64 wheels lack it)
- [Kimi-K3] Piecewise prefill CUDA graphs: run MLA attention as one eager break
- [ROCm][DSA] Enable fused k-pool DSA metadata on HIP
- [ROCm] Fix breakable prefill graph replay on non-contiguous attention outputs
- Docs
- Python not yet supported