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
- [QSA] Join gather indices instead of gathered K/V in chunk-prefill extend
- [VLM] Add SGLANG_MM_GPU_IMAGE_DECODE to keep JPEG decode off the GPU
- [VLM] Parallel, low-memory preprocessing for large Qwen-VL image batches
- 按照指导操作后，A5仍然无法拉起dsv4.1，报错如下：ValueError: The checkpoint you are trying to load has model type deepseek_v41 but Transformers does not recognize this architecture.
- [Function call] Unwrap spurious arguments/input wrapper in DeepSeek DSML parser
- [MiniMax-M3] Add native SM90 Q8K8 sparse prefill Step 1
- [Feature] [NPU] Add topology-aware CPU and NUMA affinity for Ascend worker ranks
- [DSA] Support group-scaled FP8 NoPE KV reads for CUDA TileLang
- [Bug] [NPU][MoE] AscendTPDispatcher downcasts FP32 routing weights to the hidden-state dtype
- [Feature] [NPU] Add ScatterPaKvCache as an optional KV-cache write backend
- Docs
- Python not yet supported