Your full-stack
on-device AI platform

Everything you need to build edge AI applications,
from model selection to inference, in one cohesive toolchain.

Specialize

Easily find the best model for your task and constraints, customize it to your task, and test it instantly on your device.

Best model search

Get the model most suited to your use case and device constraints.

Model library

Browse and download pre-trained LEAP model bundles from our curated library.

Fine-tuning tools

Fine-tune LFM2 (and other LEAP models) to your use-case with GPU-optimized scripts.

Apollo

Download and vibe-check LEAP models on-device with our native mobile app.

Function calling

Enable AI models to interact with external systems through structured function outputs.

Vision-enabled

Run image-to-text models on-device; available for iOS, Android, and laptop.

Package your model, integrate it into your app, and run it entirely on-device for fast, private inference.

Edge SDK

Load and query LEAP models locally on supported hardware as easily as calling a cloud API.

Model bundling service

Package any supported model or checkpoint into a bundle optimized for deployment.

Laptop support

Run LEAP models on computers via llama.cpp – optimized for AMD Ryzen™.

Get started

From model selection to on-device inference in minutes. No cloud, no guesswork — just fast, private, specialized AI.