What is laya.tools?
laya.tools is a free, independent directory of everything developers build on Laya, the open-source (Apache-2.0) System 1 decision model. Laya turns any text into typed answers — choices, scores and yes/no — in a single forward pass, running locally and supporting 100+ languages. The directory tracks 950+ projects imported daily from GitHub, npm, Hugging Face and X, covering runtimes, API servers, agents, coding tools, games and benchmarks.
Who it's for
- Developers exploring local inference — anyone who wants typed decisions from text without relying on a hosted LLM.
- Builders on Apple Silicon — people looking for MLX or Core ML runtimes that run Laya natively on their machines.
- Tooling and agent creators — those who need API servers, MCP connections, or integrations across Node.js, Rust, C++/GGUF, ONNX and OpenVINO.
Key features
Daily project imports
The directory pulls in 950+ projects every day from GitHub, npm, Hugging Face and X, so the listing stays current as new work appears.
Runtime coverage
Entries span MLX, Core ML, Node.js, Rust, C++/GGUF, ONNX and OpenVINO runtimes, giving a broad view of how Laya is being deployed.
Beyond runtimes
The catalog also includes API servers, agents, coding tools, games and benchmarks built around Laya's typed-decision output.
Independent and free
laya.tools is a free, independent directory — it exists to track the ecosystem rather than to sell a single product.
What stands out
Laya turns any text into typed decisions in milliseconds, on your own machine.
That single-pass approach is the core of the model's appeal: choices, scores and yes/no answers come back together, locally, without waiting on a hosted service. The directory reflects how quickly that idea has spread — from MLX ports playing Snake at 60 decisions per second to Core ML builds running most operations on the Neural Engine. It's a snapshot of an ecosystem forming in real time.
Worth checking out if…
You're curious about local, typed-decision models and want to see what people are actually building with them. If you work on Apple Silicon, agent tooling, or multilingual text classification, the range of runtimes and projects here gives you a practical starting point. And if you're simply tracking where open-source System 1 models are heading, laya.tools is a useful place to watch.
