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] multimodal: [MM refactor 7/9] Add a token-space strategy for Qwen VL models
- [Refactor] multimodal: [MM refactor 8/9] Check token-space Qwen output against the legacy processor
- [Refactor] multimodal: [MM refactor 9/9] Expand only the new turn of a multimodal request
- [Doc] tool-parser: Document every registered tool-call parser and test the table
- [Refactor] diffusion: Move the Nomas segment KV cache under layers/kvcache
- [Perf] [NVIDIA] GLM-5.3-Flash: Add opt-in FP8 for the BF16 KDA projections
- [Fix] hicache: Reject incomplete Mooncake KV reads
- [Perf] [NVIDIA] attention: Add opt-in optimized LiteTopK fused indexer top-k for GLM-5.2 DSA prefill on SM100
- [Perf] diffusion: Map local array spills instead of copying them
- [Perf] diffusion: Keep layerwise offload buffers cached between requests
- Docs
- Python not yet supported