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
- [Fix] kv-cache: Ignore checkpoint k/v scales when the KV cache is 16-bit
- [Perf] [NVIDIA] fp8: keep K%128!=0 block-fp8 linears on the Triton kernel under the MXFP8 view
- [Perf] openai: Cache stable chunks within long chat segments
- [Fix] [AMD] deterministic: Batch-invariant mHC pre for GLM-5.3 on ROCm
- [Fix] function_call: Accept underscore-prefixed Pythonic identifiers
- [Fix] deterministic: Put DSA k-pool prefill chunk ends on the chunked-prefill grid
- [Fix] deterministic: Deterministic k-pool top-k for the DSA indexer
- [Refactor] [Intel] weight-cache: Platform hooks for Intel XPU support
- [Perf] kimi-k3.1: Run the SP tail's shared experts on a high-priority stream
- [Perf] kimi-k3.1: Fuse o_proj with its reduce-scatter in the SP tail
- Docs
- Python not yet supported