Details
- Google AI Developers announces that Nano Banana Pro (gemini-3-pro-image) and Nano Banana 2 (gemini-3.1-flash-image) are now generally available for production use through the Gemini API.
- The thread highlights multiple community-built examples to showcase the models' capabilities, including advanced world knowledge tasks.
- Demonstrations emphasize subject consistency, showing the models can maintain coherent depictions of the same subject across different images and prompts.
- Additional examples cover style transfer, where the models apply visual styles from one image or reference to another, and robust text rendering inside images.
- A "production-ready specs" link suggests that detailed model specifications, performance characteristics, and deployment guidance are available for developers.
- Google directs developers to get started immediately with a "Start building today" link, signaling that documentation, SDKs, or quickstart resources are already live.
Impact
By pushing Nano Banana Pro and Nano Banana 2 to general availability on the Gemini API, Google strengthens its position in multimodal AI tooling for developers. The emphasis on subject consistency, style transfer, and text rendering targets practical creative and product use cases, likely intensifying competition with other image-generation APIs as teams look for stable, production-ready visual models.
