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
- Drop requests that are already too stale during a weight-update pause
- [VLM] Rust Kimi-K3 image preprocessing in sglang-mm (library mode)
- Where do CPU kernel patches go after sgl-kernel moved under sglang.kernels.aot?
- [Bug] /v1/responses API fails with 400 when stream parameter is not specified (None vs bool type mismatch)
- [Bug] [NPU] (v0.5.17)Eco-Tech/Qwen3.6-35B-A3B-w8a8 Model Served with four 910B: ValueError: Unsupported ModelSlim MoE schemes for layer mtp.layers.0.mlp.experts: W13='FLOAT', W2='FLOAT'
- feat(rust-server): support preferred sampling params
- [HCU] [CI] Add minimal PR smoke test framework
- Normalize the deprecated nsa attention backend alias to dsa
- [Fix] stream parameter None issue in Responses API _make_request
- [Bug] Qwen3.5-397B + NEXTN crashes with CUDA illegal memory access near 262144 context boundary on v0.5.16 (H800 TP8)
- Docs
- Python not yet supported