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
- [Bug] Qwen3.5-35B-A3B TP1/PP2 + NEXTN produces different greedy tokens under concurrent arrivals with deterministic inference
- [AMD][Fix] hicache: Fall back to transparent hugepages when none are reserved
- [Bug] --max-queued-requests 0 with priority scheduling crashes the scheduler on an empty queue
- [Fix] [CPU] memory: budget overestimation caused by filtering empty NUMA nodes
- [Perf] [AMD] layernorm: Use torch rms_norm in the HIP fallback without vllm
- [Feat] DeepEP: Support DeepEP v1 multi-buffer registry for target + draft MoE
- [Feat] afd: add Attention-FFN disaggregation for MoE inference
- [Perf] kda: Size the speculative ReplaySSM ring by running requests
- [Feat] linker: Add prefetch mode for the Mooncake external linker
- [Fix] mamba: Skip zero-length Mamba checkpoint reads
- Docs
- Python not yet supported