What Apple devices are compatible with On Device AI?
On Device AI works on iPhone 14 and newer, iPad mini 7th gen and newer, iPad Air 5th gen and newer, iPad Pro 3rd gen and newer, all Apple Silicon Macs, and Apple Vision Pro. Devices with 6GB or more RAM are recommended for optimal performance.
Can On Device AI work completely offline?
Yes, once you download a model, core features like voice transcription, document analysis, AI chat, Knowledge Libraries, vision models, and text-to-speech all function without an internet connection. Cloud providers are optional and off by default.
What AI models does On Device AI support?
The app supports over 170 models including Llama 3, Gemma 3, Qwen 3, DeepSeek, Phi-4, and Mistral. You can also import custom GGUF models from Hugging Face. It utilizes both llama.cpp (GGUF) and MLX inference engines for broad compatibility on Apple Silicon.
How do multi-agent AI teams function within the app?
Multi-agent teams allow you to deploy specialized AI agents that collaborate on complex tasks, such as a Researcher gathering information, an Analyst evaluating it, and a Writer producing output. Each agent can use a different model and role, all running entirely on your device.
Is On Device AI more private than ChatGPT?
Yes, On Device AI runs models directly on your hardware with zero data collection, no accounts, no analytics, and no telemetry. Your conversations never leave your device during local AI processing, unlike ChatGPT which sends data to OpenAI servers.