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
- [Perf] [NVIDIA] gemm: Enable KDA FP8 skinny GEMM for K=3072, N=5120 at M=2/4/8
- [Fix] kv-events: Retry replay batches under backpressure
- [Perf] disaggregation: Avoid O(N^2) watermark maintenance in StagingAllocator.free
- [Fix] function_call: Type the named tool_choice item schema as an object
- [Perf] [AMD] gdn: Add an aiter backend for linear-chain target verify
- [Test] attention: Guard Muse Spark fused KV store writes
- [Perf] qsa: Pass output_size to repeat_interleave in the graph speculative layout
- [Fix] hicache: Correct DCP L2 transfer-byte accounting
- [Refactor] memory-pool: Bind HiCache entries to device pools by name
- [Refactor] memory-pool: Move the NVFP4 host budget into its host module
- Docs
- Python not yet supported