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
- [diffusion] comfyui: add SGLDCacheOptions for per-run Cache-DiT
- [Bugfix] DeepSeekV3 streaming: emit every complete tool call in a buffered backlog
- [diffusion] fix: OpenAI API hardening and runtime utils/pipeline bug fixes
- Add --speculative-dcp-size (default 1, > 1 rejected)
- Fail fast when dtype=float32 is used with the flashinfer attention backend
- [Bug] `--enable-deterministic-inference` does not make hybrid Mamba2 models (granite-4.0-h) batch-invariant
- [diffusion] Fix dLLM running timeout between denoising rounds
- [Diffusion] Capture Cosmos3 GEN MLPs with breakable CUDA graphs
- [DSA] Reuse metadata kernels across batch sizes
- [Perf] Enable native MXFP4 routing and CuTeDSL fusion for MiMo
- Docs
- Python not yet supported