Ollama Mobile: AI Chat

Use your own Ollama models from your phone. Connect to a local server, a remote one, or Ollama Cloud. Chat, generate images, and search the web. Data stays on device.

Ollama Mobile is a native client for Ollama. Connect to a local server on your own network, a self-hosted remote server, or Ollama Cloud, and chat with your models from your phone.

Your phone is the client, not the model host. The app sends your prompt to the Ollama server you choose, that server runs the model, and the reply streams back to your screen. Nothing multi-gigabyte downloads to your phone.

WHAT YOU CAN DO

• Streaming chat with live token rendering and a collapsible thinking channel
• Markdown rendering for headings, lists, tables, and inline formatting
• Image generation with a full-screen, pinch-to-zoom viewer
• Image attachments for vision-capable models
• Web search through tool-calling on Ollama Cloud
• Model catalog to list, inspect, pin, and pull models from any endpoint
• Prompt library with built-in prompts plus your own
• Multiple endpoints in one app: local, remote, and Ollama Cloud, each with a test-connection check
• Message actions: copy, edit, resend, regenerate, and export a chat as text

PRIVACY BY DESIGN Ollama Mobile collects nothing. There are no accounts, no analytics, and no third-party SDKs. Your chats, prompts, endpoints, and settings stay on your device, and API keys are held in the device Keychain. The app talks only to the servers you configure. Connect to a server on your own network and your conversations stay on that network. If you lose the device, the data is gone, by design.

WHAT YOU NEED An Ollama server to connect to: Ollama running on a computer you control, a remote server you host, or an Ollama Cloud account. Real conversations need a reachable server, because the model does not run on the phone.

Ollama Mobile is an unofficial, independent client. It works with Ollama but is not made by, endorsed by, or affiliated with the Ollama project.