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
- [Test] moe: Pin the Waterfill algorithm end to end
- [Perf] [NVIDIA] fused_moe: Add H20 configs for GLM-5.3-Flash TP8/TP4
- [Fix] flashinfer: Keep sliding-window layers off the ragged prefill wrapper
- [Fix] flashinfer: Reuse CUDA-graph wrappers when a batch size is captured twice
- [Feat] volundr: Add Agens Volundr model support
- [Perf] DSv4/DSv4.1: Skip the pre_hc_head CP gather when DSpark captures aux hidden states
- [Feat] unified_cache: Publish direct-UMBP residency as host-tier KV events
- [Perf] mooncake: Group and parallelize direct linker layer-wise loads
- [Fix] ngram: Reject capacity <= max_trie_depth instead of aborting the scheduler
- [Perf] function_call: Eliminate O(N^2) streaming bottleneck in BaseFormatDetector
- Docs
- Python not yet supported