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
- [Feature] Expose the effective `max_running_requests` (after the mamba/linear-attention state-cache cap) in `/get_server_info`
- [Kernel][MoE] Add a CUTLASS MXFP4 x BF16 grouped-GEMM MoE backend for SM90, improving long-context prefill by up to 20%
- [Feature] Move shared cache transfer types out of hicache_storage.py
- [Bug] MiniMax H3 GGUF text encoder fails loading folded Conv3D patch embedding
- Expose effective max running requests in server info
- [KDA] Defer accepted-state materialization for SM90 Triton verify
- [Bug] SM120 grouped FP8 DeepGEMM weight preparation skips UE8M0 requantization
- [Bugfix] Infer the draft path for Qwen3.5 bundled MTP
- fix(openai): honor include_reasoning=false on chat completions
- [core] Make SWA sizing policy explicit and separate pool costs
- Docs
- Python not yet supported