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
- [mem_cache] soft retention: priority decay via retention_seconds
- [diffusion] Add FlashInfer trtllm-gen skip-softmax (BLASST) attention backend for DiT
- [MoE Refactor] Add torch_native MoE runner backend
- docs(cookbook): migrate Kimi-K2.6 to the config-driven template
- docs(skill): fix cookbook-skill gaps
- [AMD][Perf] Fuse QK-norm + 3D mRoPE + KV cache store into single aiter op for Qwen3.5-397B-A17B-MXFP4 on HIP
- [mem_cache][12/N] refactor: stop allocators round-tripping through the KV pool
- [router] cache_aware_zmq: load-aware selection from engine LoadSnapshot
- [AMD][Perf] Fuse QK RMSNorm + 3D mRoPE + KV-cache store into single aiter op for Qwen3.5-397B-A17B-MXFP4 (TP=2, ROCm/aiter) on HIP
- perf(sgl-router): shard cache-aware-zmq radix tree to remove read-vs-write lock contention
- Docs
- Python not yet supported