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
- [Bugfix] Fix idle detection for unified cache external linker
- [Bug] Cold TokenizerWorker can hang on pause/continue because its receive loop is not initialized
- [MoE] Add flashinfer_cutedsl_fp8 runner backend
- Fix cold tokenizer worker pause and continue initialization
- [Performance] Cap SWA windows per radix path and reserve Mooncake loads
- [Performance] Batch Mooncake hybrid-pool metadata and puts
- [HiCache] Retry failed storage backup batches and expose terminal failures
- [Attention] Honor custom masks beyond causal query tiles in Triton extend
- [Fix] Drop the shadowed copies of the lm_head quant-method helpers
- [Bug] Janus-Pro CLIPVisionTower image normalization cannot run: Normalize is shadowed, and the transform calls torch.nn.functional.normalize
- Docs
- Python not yet supported