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] Plan residency as one pool on unified memory (GB10)
- [AMD]Add opt-in DSV4 ROCm attention multi-stream overlap
- [Fix] Make test port derivation robust to any CUDA_VISIBLE_DEVICES value
- docs(cookbook): compute MiniMax-H3's --sp-degree from ulysses x ring, not the world size
- fix: release LoRA adapter on request abort
- Add --offload-embedding-to-host: keep the input embedding table in pinned host memory
- Support --language-model-only for Qwen3.5 VL checkpoints
- [Bug] --enable-linear-replayssm forces no_buffer, which degrades mamba prefix caching and inflates TTFT up to 4.7x
- [Bug] Null optional config sections are treated as populated
- [CPU][Spec] Support DFlash target-verify for hybrid GDN models
- Docs
- Python not yet supported