IMAGE MODEL by Google DeepMind Gemini 2.5 Flash Image

Nano Banana

Google's precision image model — known for one-shot accuracy, multi-image editing up to 4 references, and near real-time speed. Built for branding, e-commerce, and campaign workflows.

References

Up to 4

Speed

Near real-time

Accuracy

One-shot

Google

Nano Banana is available exclusively through ImagineArt, integrated into both the AI Image Generator and the Image Editor. ## Capabilities Maintains subject identity, lighting, and style coherence across multiple outputs — ideal for brand campaigns and product series. Delivers high-quality results at near real-time speeds — efficient for quick iteration and large-scale projects alike. Adapts to text-to-image generation, guided image editing, and multi-reference design creation across different workflows. 4 Ii Vo3t0 Max 1300x1300

## How to use Go to the **ImagineArt AI Image Generator**. From the model dropdown, choose **Google Nano Banana**. Write your text prompt and upload up to 4 reference images for richer guidance. Click **Generate** and wait for the model to process your request. 2 H J Wj DOO Max 2000x2000

Go to the **ImagineArt Image Editor**. Click **Settings** and select **Google Nano Banana** from the model dropdown. Apply edits or transformations directly to your uploaded visuals. Use **Visual Prompts** to make further adjustments, add annotations, or overlay images on your base image. Go to the **ImagineArt AI Video Generator**. Choose **Kling 2.1** or your desired video model from the dropdown. Set the resolution to **1080p**. Use **Google Nano Banana** to edit and refine the start and end images before generating the video. ## Prompting tips Nano Banana is optimized for one-shot accuracy — precise, structured prompts produce the best results. 1. **Be specific** — Define the subject, background, product, and style in one prompt. Example: *"A man in a modern kitchen holding a coffee mug, styled like a product ad."* 2. **Use multiple references** — Upload images of the subject, product, and setting together to maintain consistency across outputs. 3. **Guide the mood** — Include terms like *"cinematic"*, *"flat lay"*, or *"studio lighting"* to shape composition and tone. 1 U1xks1l Max 1400x1400

### Example prompts > A fluid gender albino person inside a vintage car with bright yellow chrysanthemums spilling out through the windows, holding a baby blue handbag. The scene blends nature and fashion with a dreamlike atmosphere, shot in a cinematic and analog photo style. > A sleek modern kitchen with stainless steel appliances and a marble countertop, featuring a woman in a red dress holding a glass of wine. The scene should reflect contemporary elegance, with soft lighting highlighting the product and surroundings. ## Compare models | Model | Visual style | Strengths | Best for | | ---------------------- | -------------------------- | -------------------------------------- | ----------------------------------- | | **Nano Banana** | Photorealistic, adaptive | One-shot accuracy, multi-image editing | Image editing, branding, e-commerce | | **Flux Kontext** | Artistic, flexible | Strong stylization, precision control | Creative edits, experimental work | | **Seedream 4.0** | Photorealistic, consistent | Up to 6 references, native 4K | Complex campaigns, high-res assets | | **ImagineArt 1.5 Pro** | Realistic, cinematic | Native 4K, composition control | Posters, typography-based designs |