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
- fix: resolve QKV layout inference for MiniMax-H3 checkpoints
- [DSV4.1] DeepSeek-V4.1 support prefill CP
- [Feature] Extend Candidate Compact Scoring to DeepSeek-V4.1 DSpark Target Verification on Hopper
- [NPU][GLM5.2] support dsa pcp with interleave & zigzag
- [PD] Add opt-in HRRN decode preallocation ordering
- [SMG] Honor concurrency-only admission and retain permits through streaming
- Fix Sarvam MLA model for FlashInfer TRT-LLM MoE runner and CUDA graphs
- [diffusion] lora: revalidate FLUX.2 FP8/NVFP4 fast paths after LoRA attach
- [Quant] Route hybrid NVFP4 MXFP4 draft experts through the CUTLASS MoE method on SM90/SM120
- [Fix][DSV4.1][SM90] Avoid materializing the full decode indexer slot map
- Docs
- Python not yet supported