Skip to main content
Wally runs frontier-scale open models on our hardware, reached with one command or an OpenAI-compatible API. The same commands run smaller models on your own machine when you want them to.

Wally Cloud

We host the model. Call it like the OpenAI API, from the CLI or your own code.

On-device SDK

Ship a model inside your app, on the user’s hardware, offline.

Next steps

Cloud quickstart

Install, sign in, and run a hosted model.

First API request

A curl you can paste into a terminal.

Integrations

Point opencode, or any OpenAI-compatible client, at a model.

Run one locally

Smaller models on your own machine, no account.

Pick an SDK platform

Swift

iOS, macOS

Kotlin

Android

React Native

iOS, Android

Flutter

iOS, Android

Web

Browser

Electron

Desktop