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
- [Bugfix] Close received POSIX fds when exchange fails
- [rust-processor] Keep parity fixtures on Hugging Face
- [DSv4.1] Score candidate blocks in paged decode consumer layers
- [DSv41] Engram table backends: peer-visible device shards, host tables, a gather view
- [Bug] [diffusion] /v1/images/edits accepts a mask upload but silently ignores it
- [RFC] Native worker batching for ComfyUI diffusion requests
- Speculative decoding crashes server on first non-greedy request: flashinfer verify-sampling op JIT fails (bundled cccl vs pip cu13 toolkit) and kills the process tree
- [diffusion] Pass model-declared image fields through /v1/images/edits
- [Fix] Use CuTeDSL MLA for draft prefill and extend
- [Quant] Repack NVFP4/MXFP4 Marlin MoE experts in place (peak load memory 1.56 -> 0.39 GiB per layer)
- Docs
- Python not yet supported