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
- Extend dLLM volatile-block cache fix (#35270) to swa/mamba/unified/cpp radix caches
- [Feature] Enhanced KV Cache Observability: Full-Lifecycle Metrics and Diagnostics
- [AMD] Implement DeepSeek V4 DCP on ROCm
- refactor(npu): centralize stream management (#1104)
- test(disaggregation): add PD combo scenario unit tests (#1105)
- docs(npu): consolidated NPU graph mode usage guide (#1101)
- [Bug] When using hicache to enable L2/L3 storage in DeepSeek V4, token loop repetition occurs very frequently
- Support x-request-id header for rid pass-through
- [diffusion] MiniMax-H3: shard QKV weights across TP and Ulysses
- [PD] Support PP x PD transfer for DSv4
- Docs
- Python not yet supported