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] responses: Merge a commentary preamble with its tool calls when replaying history
- [Feat] cp: Add Qwen3.5/Qwen3.8 prefill with Symm and MegaMoE
- [Perf] [AMD] GLM5.3: Allow an FP8 E4M3 KV cache for Triton sparse attention on gfx950
- [Fix] disaggregation: Fail a prefill KV transfer whose decode never completes it
- [Fix] disaggregation: Advance a Mamba request's unpublished KV under a write-through cache
- [Feat] Add source-bound staged execution
- [Bug] SAIL/PPU memory capacity query incorrectly invokes nvidia-smi and warns about NVIDIA drivers
- [RFC] Add NPU RLHF Lifecycle Tests
- [Fix] dp-attention: Preserve original forward mode in prefill graph replay
- [Fix] [SAIL] utils: Use device properties for memory capacity
- Docs
- Python not yet supported