Ollama has officially launched a desktop application for both macOS and Windows, making powerful AI models more accessible to a wider audience. This new graphical user interface (GUI) eliminates the need for command-line interactions, allowing users to easily download AI models and chat with them directly. 

Enhanced User Experience with File and Image Support. 

The Ollama desktop app significantly improves the user experience by introducing drag-and-drop functionality for text files and PDFs. Users can now effortlessly feed documents to AI models for tasks like summarization or analysis. The app also boasts the ability to read code files, empowering developers and coders with AI assistance for comprehension and documentation generation.

Multimodal AI and Flexible Context Length

A key new feature is multimodal input support, enabling users to send images to compatible AI models such as Google DeepMind's Gemma 3. This expands the types of interactions and applications for local AI. Furthermore, users can adjust the context length within the settings, allowing the AI models to process larger documents, though this may impact memory usage. For those who prefer the traditional approach, Ollama's CLI tools remain available on GitHub.