Ollama launches a new desktop app for macOS and Windows with image and code file support

Ollama has launched a desktop app for macOS and Windows, bringing its AI models to a graphical interface that doesn’t require command-line use. Users can download models and chat with them by typing prompts, making it more accessible for a broader audience.

The app supports drag and drop for text and PDF files, with adjustable context length in settings to handle larger documents. Increasing this setting may result in higher memory usage. It also reads code files, allowing models to assist with comprehension and documentation tasks.

Multimodal input is now supported, letting users send images to compatible models like Google DeepMind’s Gemma 3. For those who prefer the command line, Ollama’s CLI tools are still available on GitHub.