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
- cpu: Support DFLASH and DSPARK speculative decoding
- [Fix] Treat whitespace-only content as empty in reasoning fallback
- [Fix] Flush buffered Harmony output at stream end
- [Fix] Reject unsupported xgrammar pattern and length combinations
- [Fix] Respect Spark2.5 attention gate activation mode
- [Fix] Handle Ollama chat inputs and context limits
- [AMD][Fix] Exclude added-vocabulary tokens from synthetic multimodal prompts
- [DCP] flashmla: reject decode context parallelism
- [NPU] Fix MTP draft loading for ModelSlim W8A8 checkpoints: dequantize int8 MTP weights
- [KDA] Optimize ReplaySSM decode with pre-decayed keys and circular checkpointing
- Docs
- Python not yet supported