Installation

Swift Package Manager (Xcode)

Open your project

Open your project in Xcode 26 or newer.

Add package dependency

Go to File → Add Package Dependencies…

Enter repository URL

Enter https://github.com/RunanywhereAI/runanywhere-swift

Select version

Choose Up to Next Major from 0.20.24.

Choose products

RunAnywhere is required. Add the backends you need.

Package.swift

dependencies: [
    .package(url: "https://github.com/RunanywhereAI/runanywhere-swift", from: "0.20.24")
]

Then add the products your target uses:

Package.swift

targets: [
    .target(
        name: "YourApp",
        dependencies: [
            .product(name: "RunAnywhere", package: "runanywhere-swift"),
            .product(name: "RunAnywhereLlamaCPP", package: "runanywhere-swift"),
            .product(name: "RunAnywhereONNX", package: "runanywhere-swift"),
        ]
    )
]

Depend on runanywhere-swift, not on the runanywhere-sdks monorepo. It is the Swift-only distribution generated from that monorepo: a few MB instead of a few hundred, and it carries the generated proto sources the monorepo no longer commits. Its tags are bare semver with no v prefix, which is what from: expects.

Available products

Product	Description
`RunAnywhere`	Core SDK. Required for everything.
`RunAnywhereLlamaCPP`	llama.cpp backend: LLM and VLM on GGUF models
`RunAnywhereONNX`	Sherpa-ONNX backend: STT, TTS, VAD, embeddings
`RunAnywhereMLX`	Apple MLX. Physical iOS device or native macOS only.
`RunAnywhereNeuRT`	Apple Neural Engine backend

Include only what you need. Each backend carries native libraries, so each one you add grows the app.

Requirements

Item	Minimum
Platforms	iOS 17.5, macOS 14.5
Xcode	26, with Swift 6.2
Disk	Several GB for SDK artifacts and downloaded models
Device	Apple Silicon recommended. MLX needs a physical device or native macOS.

Register backends before initializing

The plugin registry has to know which engines exist before the first model load. Register first, initialize second, or loading fails with “No provider could handle the request”.

import RunAnywhere
import RunAnywhereLlamaCPP
import RunAnywhereONNX

LlamaCPP.register()
ONNX.register()

try RunAnywhere.initialize(apiKey: "", baseUrl: "", environment: .development)

Troubleshooting

Symptom	Fix
Missing XCFramework errors	File → Packages → Reset Package Caches, then resolve again
Package resolution failures	Same: reset caches and resolve
”No provider could handle the request”	A backend was not registered before `initialize()`
MLX unavailable	Use a physical device or native macOS; MLX does not run on the simulator

Swift SDK

Kotlin SDK

React Native SDK

Flutter SDK

Web SDK

Electron SDK

Swift Package Manager (Xcode)

Package.swift

Available products

Requirements

Register backends before initializing

Troubleshooting

​Swift Package Manager (Xcode)

​Package.swift

​Available products

​Requirements

​Register backends before initializing

​Troubleshooting

Swift Package Manager (Xcode)

Package.swift

Available products

Requirements

Register backends before initializing

Troubleshooting