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
- [rust-server] fix rust response metadata parity
- [Bug] OpenAIServingChat._get_reasoning_from_request Ignores the Inkling "reasoning off" Signal
- [Bug] OpenAIServingChat.apply_reasoning_enabled Does Not Mirror the minimax-m3 Read-Side Rule
- [Bug] get_hf_text_config` erases the text sub-config's `dtype` on the `thinker_config` path
- [Bug] sglang.Engine `chunked_prefill_size=-1` derives a negative `mem_fraction_static` and the engine fails to start
- [Bug] sglang.Engine crashes with `KeyError: torch.float32` when the documented `dtype="float32"` is used
- [Docs] Add Qwen3.5 B300 NVFP4 agentic recipes to the selector / [Docs] 在 Qwen3.5 选择器中添加 B300 NVFP4 Agentic 配方
- [PD] Fix the early-send wait event on non-CUDA devices and in nixl
- [Fix] Run:ai streamer: copy CPU-path tensors out of the staging buffers
- [Docs] Add Qwen3.5 GB200 Dynamo recipe selector / [Docs] 添加 Qwen3.5 GB200 Dynamo 配方选择器
- Docs
- Python not yet supported