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
- [Feat] disaggregation: Pool-declared PD KV layout
- [Feat] mem_cache: Platform hook for the page_size == 1 allocator
- [Feat] platform: Let a graph runner that owns prefill be built on PD prefill servers
- [Fix] disaggregation: Gate the decode retraction host-pool default on host-memory support
- [Fix] constrained: Keep grammars that allow no token from crashing the server
- fix(gemma4): index streamed tool calls by response order, not tool po…
- Fix flashinfer_megamoe transport rows undeclared sum group error
- Fix deepseek-recipe rendering: add tool namespace and action field support
- [Fix] beam-search: Correct mixed-width admission and reduce decode overhead
- [Perf] tokenizer_manager: Reduce incremental streaming metadata scans
- Docs
- Python not yet supported