v0.20.2

Ollama, the massively popular open-source platform for running large language models (LLMs) locally, has shipped a subtle but user-focused update with version 0.20.2. The change, implemented by core maintainer @jmorganca, modifies the application's default home view. Previously, opening Ollama would land users on a screen to select and launch a model. Now, it opens directly to a fresh, empty chat interface. This prioritizes the core action of interacting with AI over administrative setup, streamlining the workflow for developers and enthusiasts who frequently use the tool.

While a minor point release in the changelog, this update reflects Ollama's commitment to refining the user experience for its 167k GitHub stars-strong community. The change is immediately noticeable upon launching the updated app, removing a friction point for power users who run models like CodeLlama or Phi-2 for development tasks. It signals a focus on the application layer as the ecosystem matures, moving beyond pure infrastructure to polish the daily interaction loop. The community reaction on GitHub was positive, with 23 reactions including 14 thumbs-up and 16 celebratory emojis, indicating approval for this quality-of-life improvement.