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 NIXL bootstrap handling for foreign traffic
- [Diffusion] Support SM12.x GPUs (RTX PRO 6000 Blackwell / RTX 50xx / DGX Spark GB10)
- [Bug] sgl-model-gateway router rejects /v1/responses requests with tool type "custom" — openai-protocol crate out of sync with Python protocol.py
- [Feature] Investigate CUTLASS 4.6 dynamic Epilogue Fusions and IKET profiling
- [Bug Fix][Compilation] Use a deterministic computation_graph dump filename to stop unbounded per-startup accumulation
- [DeepSeek V4] Disable YaRN for uncompressed attention layers
- fix: add exist_ok=True to qwen3_asr AutoConfig.register
- Fix mamba conv state dtype defaulting to bfloat16 for FP16 hybrid linear-attention models
- [Feature] Enable a16w4 MXFP4 MoE on the aiter fp4_bf16 FlyDSL path
- Reduce tensor-parallel logits communication with a rank-0 gather
- Docs
- Python not yet supported