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
- [Tracking][Mori] PRs for Community Review
- [diffusion] quantization: packed Q/K/V projection for convrot_int8's comfy_kitchen backend
- [Fix] epd: Reject dequeued encoder requests when stopping during coalescing
- [Fix] compressed-tensors: Serve NVFP4 W4A4 linears via FP4 Marlin when --fp4-gemm-backend marlin is set on SM100
- [PD] Support dp/dp attention for PD role switch
- [DSA] Support flashmla_sparse_q8 prefill backend on CUDA
- [Fix] cuda_graph: Wait for in-flight layer-wise cache loads before prefill graph replay
- [RFC][Simulator] Support Closed-Loop Agentic Workloads
- [Fix] speculative: Validate QSA top-k before loading weights
- [Fix] diffusion: keep host-resident embedding scales on CPU
- Docs
- Python not yet supported