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
- Refactor: Deprecate Prefill CP V1
- [AMD] Create the hipBLAS handle before CUDA graph capture
- [Bug] DeepSeek-V4 returns an empty response when the final message has role=system
- [Bug] `mm_input_embeds` is corrupted before the draft model reads it
- add mxfp scheme support for AR
- [Bug] Model generation returned no output. Error from scheduler: Error executing request bda883a6-0e33-42d6-afa2-b3038f37ba2e: Forward context is not set. Please use `set_forward_context` to set the forward context.
- [Perf] Keep RadixKey prefix matching on the array('q') memcmp path
- MXFP8 x BF16 MegaMOE support
- [Feature] Support Kimi-K3 FlashInfer MXFP4 MoE on SM120
- [Playground] Verified cell: dgx-spark / default / nvfp4 / undefined / single
- Docs
- Python not yet supported