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][AMD] Enable the bit-exact Triton SiLU-mul fusion on HIP
- [AMD] Support DeepSeek-V4 context-parallel prefill two-batch overlap
- [NPU] Fix Ascend CPU affinity using runtime device topology
- [AMD][Fix][MiniMax-M3] Choose the aiter FP8 cache mode from the pool's KV scales
- Feature request: optional compact readout for PPLX decision checkpoints — skip the materialized vocabulary head (~2.37 GiB) and read the native 255-row readout directly
- [Perf] [AMD] batch-invariant: Tune the bf16 matmul for gfx950
- [Feature] Opt-in compact readout for PPLX decision checkpoints
- [Fix] Validate sampling bounds and preserve chat output limits
- [AMD][Quark] Choose the activation precision for MXFP4 dense linears on ROCm
- [AMD][Kimi-K3] Let Quark MXFP4 shared experts join the fused MoE front
- Docs
- Python not yet supported