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] diffusion: Reject requests and report unhealthy after an unrecoverable device error
- [Feature] jit: Probe-once fail-fast when the toolchain cannot compile anything (glibc 2.42 rsqrt clash)
- [Perf] [AMD] dsv4: Gather the compressor kv_score in bf16 under prefill CP
- [Feature] Add GLM-5.3-Flash NVFP4 KV cache on SM90 with FA3
- [Feature] Offline batch runner for OpenAI batch files (follow-up to #7427)
- [Perf] hicache hf3fs: Copy and submit per page in usrbio batch_write
- [Fix] qwen3_next: dequantize GGUF weight for shared embedding/head
- [Fix] [NVIDIA] deepseek-v4.1: Fix prefill CP (interleave) crashes on SM90/SM100/SM120
- [WIP][Refactor] PD+PD: Remove ring-consensus for release at prefill
- [Fix] [DSV4.1] mHC: don't specialize split-K Triton kernels on row count
- Docs
- Python not yet supported