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
- [Refactor] memory-pool: Report device pool roles in buffer info
- [Feat] [Intel] disaggregation: Support Intel XPU in the EPD multimodal encoder path
- [Fix] moe: Keep FP8 TRT-LLM in-kernel routing for padding-only TopK bounds
- [Bug] rust-server: gather_server_endpoints can advertise a loopback or unroutable address to other nodes, and only warns
- [Perf] [NVIDIA] muse-spark-open: Fuse all-gather + Norm + Quant
- [Perf] R3: Gather DeepEP routed-expert ids once per forward (3/3)
- [Perf] [NVIDIA] muse-spark: Build the routed-MoE routing metadata on router stream
- [AITER-Upgrade] PR readiness gfx1250 (MI45x)
- [Fix] [AMD] hicache: Book explicitly sized host pools against one host memory snapshot
- [Perf] [NVIDIA] kernels: Add Blackwell SM100 CuTe DSL NVFP4 DualGEMM+SwiGLU, PDL-safe GEMM, Grouped GEMM, and Fused MoE
- Docs
- Python not yet supported