Run Laya anywhere
Laya Runtimes & Local Deployment
Choose the runtime that matches your language stack, hardware, memory budget, and serving model. These pages track both upstream and community implementations.
Laya Python / PyTorch
The upstream reference implementation and Router. Start here for the canonical API, model checkpoints, fine-tuning, and cross-platform inference.
Laya MLX for Apple Silicon
Community-native MLX runtime for fast local typed decisions on M-series Macs without PyTorch.
Laya Node.js / TypeScript via ONNX
Run Laya from Node.js and TypeScript using ONNX Runtime, with the same typed-decision request and response shape.
Laya MPS on macOS
A memory-conscious local runtime and HTTP server for Apple GPU / MPS, with multiple memory modes and benchmark tooling.
Laya Local Serving with Arbiter
Serve Laya behind a Jev-compatible HTTP API with checkpoint routing, batching, metrics, a Playground, and NVIDIA / Apple Silicon recipes.
Laya in Rust: Official SDK, sys1 and ollaya
Run Laya in Rust in-process with the official laya-onnx SDK on ONNX Runtime, or as a Jev-compatible server with the community projects sys1 and ollaya.