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] Forward scheduler SIGQUIT through the DP controller
- perf(unified-memory): avoid repeated source scans in float compaction
- refactor: openai api to use general generation path
- [ROCm] Restore partial DP-attention correctness and CUDA-graph safety
- [Bugfix] Discover profile traces written with a profile_prefix
- [Bugfix] Ignore PREBUILT batches in profile V2 stage mapping
- [MLX] Pass through non-q4/q8 MLX quantization_config bit widths
- [HiCache] Yield to storage threads on every no-batch loop iteration
- [Spec][DSpark] Use live gamma for DSV4 confidence
- [Spec][CUDA Graph] Capture compact verify token keys
- Docs
- Python not yet supported