.svg){kind=spacer}
Details
- Google for Developers announced that Nano Banana Pro and Nano Banana 2, its latest image generation models, are now generally available on the Gemini API.
- Nano Banana 2, corresponding to Gemini 3.1 Flash Image, is optimized for high speed, lower cost, and high-volume image workflows.
- Nano Banana Pro, corresponding to Gemini 3 Pro Image, targets complex creative pipelines and professional use cases that require higher precision and control.
- The models offer subject consistency across up to five characters, accurate text rendering with localization, support for 4K upscaling and custom aspect ratios, and up to 14 reference image inputs.
- Both models can ground generations with Google Search, improving factual accuracy in generated visuals and text elements.
- Google highlights early adopters such as Spline (3D mesh generation) and Krea (real-time editing), showing how partners are already using the models for advanced creative and interactive workflows.
- A linked Google blog post provides deeper technical details and integration guidance for developers building production applications on top of the Gemini image stack.
Impact
By moving Nano Banana Pro and Nano Banana 2 to general availability on the Gemini API, Google strengthens its position in the competitive AI image generation market against providers like OpenAI and Stability AI. The combination of speed, subject consistency, and Google Search grounding is likely to attract developers building production-grade creative tools, marketing systems, and 3D or real-time editing applications, potentially accelerating enterprise adoption of Gemini-based image workflows.
