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] minimax-m2: Keep values of nullable parameters and survive non-finite numbers
- [Perf] [AMD] hicache: Default the all-layer MLA transfer ops to the MLA block quota
- feat: port Muse Spark Open support to sglang-miles
- [Perf] hicache: optional minimum tokens per write_back allocation-driven eviction
- [Fix] model_loader: Bound CPU checkpoint retention in multithreaded PT loading
- [Fix] mem_cache: Fix HiCache load-back rollback and pin handoff
- [Perf] [NPU] embeddings: Fuse token remapping and vocabulary mask
- [HiCache] Storage touch hook and prompt_cache_options.ttl retention hint
- [Feat] [Intel] diffusion: SageAttention V3 for XPU self-attention
- [Feat] [Intel] diffusion: online MXFP8 on XPU
- Docs
- Python not yet supported