OpenBrowse

Google Gemini

Use Gemini 3.1 Pro, Flash, and experimental models.

OpenBrowse supports Google's Gemini models via the Gemini Developer API. Gemini 3.1 Pro and Gemini 3.1 Flash offer massive context windows (up to 2 million tokens), which is incredibly useful when asking the agent to read and process very long web pages or analyze multiple tabs simultaneously.

Setup

  1. Go to Settings → Models.
  2. Select Google.
  3. Enter your Google Gemini API key.
  4. Choose a default model, like gemini-3.1-pro-preview.

Thinking Mode

OpenBrowse supports Google's thinking models. When selecting these models, you can enable thinking mode to improve complex reasoning capabilities.

On this page