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
- [Model] Boogu-Image: reference-image edit (TI2I / I2I) pipeline
- perf(kimi k3): eliminate redudant comm in mixed tp and ep
- [RFC][sgl-router] Recoverable KV Placement State with Snapshots and Event Replay
- Fix DSPARK SM120 decode dispatch for non-instantiated topk widths
- [Bug] `max_running_requests` (4096) and `cuda_graph_max_bs` (32) defaults are derived independently; crossing the graph ceiling is an absorbing state
- Rainj me/rust server pd lb
- docs(cookbook): add Community Configs section for contributed recipes
- feat(kimi-k3): run native MXFP4 MoE with humming for hopper
- [perf] Quantize TRTLLM MLA prefill Q once for chunked-prefix attention
- [Fix] Support transformers >= 4.57 configs for DeepSeek V4
- Docs
- Python not yet supported