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] deepseek: Remap block-FP8 weight_scale for stacked gate_up params
- [Feat][PoC] beam_search: Add paged KV support and item-catalog constraints for GR
- [Perf] linker: Build Mooncake grouped load range metadata with numpy
- [Perf] linker: Resolve slot locations before the forward fence
- [Perf] linker: Complete landed Mooncake layer groups between submits
- [Perf] linker: Fill the restorable-prefix mask with one vectorized store
- [Perf] linker: Load DSpark draft KV with the last transfer layers
- [Fix] linker: Cover published prefetch KV with its request's session
- [Perf] linker: Build every Mooncake layer group's range reads before the forward fence
- [Perf] SANA-Video: Compact Masked Cross Attention tokens with Dual Width BCG capture
- Docs
- Python not yet supported