Ollama 1.0: first stable release with multimodal, tool calling, and Windows GA

Ollama is the most widely used tool for running AI models locally. With version 1.0, it exits the experimental phase and brings important new features. You can finally send images to models that support them, like LLaVA or Llama Vision. Models can natively call external functions and tools. You can generate embeddings for your documents. And for the first time, Windows is officially supported as a first-class platform — no longer just macOS and Linux.