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
- [AMD][ROCm] Keep cos_sin_cache fp32 on HIP for fused QSA indexer kernel
- [Fix] Feed the relayed last token to penalizers under the overlap scheduler
- [Perf] Use the fused low-ratio compressor on Hopper
- [CPU] Add FP8 quantization support for compressed-tensors paths
- [CPU] Add INT8 quantization support for per-token and compressed-tensors W8A8 paths
- [diffusion] feat: add --async-output-save to overlap output saving with the next request
- NIXL: Use stride desc API
- Make prefill shared-read completion explicit and default-on
- [Perf][mem_cache] Persistent lazy eviction heap for the UnifiedRadixCache Full component (bit-identical order)
- [MiniMax-M3] Instantiate the decode block top-k in register buckets
- Docs
- Python not yet supported