From the archive
Google opens Gemini Pro through AI Studio and Vertex AI
Gemini Pro becomes available through a developer API, with AI Studio for prototyping and Vertex AI for Google Cloud deployments.
Google opened Gemini Pro to applications through the Gemini application programming interface (API). The release also introduced Google AI Studio as a browser-based place to test prompts and create an API key.
Two ways to build with Gemini
AI Studio provides a short path from trying a prompt to exporting code for an application. Vertex AI places Gemini inside Google Cloud's managed AI platform, with cloud account controls and additional deployment and data-management features.
- Gemini Pro accepts text and returns text.
- Gemini Pro Vision accepts text and images and returns text.
- The developer tools include support for function calling and chat.
The initial Gemini Pro text release used a 32,000-token context window.
Access and data handling
Google offered limited free access at launch. Its announcement said AI Studio's free-quota inputs and outputs could be examined by trained reviewers, while Google Cloud customer inputs and outputs were not used to train its models. That distinction matters when choosing a route for confidential application data.
The Google DeepMind profile explains Gemini and direct developer access. The separate Google Cloud AI profile covers managed cloud deployment.