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
- [NPU][MiMo] Fold attention_value_scale into o_proj weights
- [DeepSeek-V4] Enable fused and overlapped mHC stats for small SM90 decode batches
- [NPU]fix npu hicache ci oom
- Support a reduced draft vocabulary in rejection sampling
- fix(inkling): return 400 instead of 500 for unfetchable/undecodable multimodal input
- fix: harden detokenizer against custom tokenizers with stale/edge-case bugs
- [Test] Fix undefined variables in FlashMLA tests and AMD ROCm benchmarks
- [NPU][LoRA] Update for follow-up with changes in sgl-kernel-npu
- [Disagg] Support heterogeneous TP for hybrid SWA KV transfer
- perf(hicache): make batch-copy threshold env-overridable for small pages
- Docs
- Python not yet supported