The latest Ollama release, v0.34.0, introduces direct integration with ChatGPT Desktop, letting you use your locally hosted Ollama models without leaving the ChatGPT interface. Once set up through the Ollama app on macOS, you can keep your existing workflow while running open models side by side.
This update also improves structured output performance on Apple Silicon, making response generation snappier for users with M-series chips. Additionally, the release adds support for OpenAI-compatible client tool search and response compaction, which streamlines how compatible applications discover tools and manage context length.
