Nano AI is an online creation workbench for image and video generation, integrating multiple mainstream AI models into a single account and a single credit system.
Purpose
Nano AI is used to turn text descriptions or reference images into visual content, making it suitable for creators who need to quickly generate marketing materials, social media images, concept art, or perform image editing. It emphasizes "choosing models by results," with each model labeled with its areas of expertise and actual credit cost, helping users make decisions based on needs rather than names.
Main Capabilities
- Image generation and editing: Supports 17 image models, including Nano Banana 2.1, GPT Image 2.5, Seedream 5.0, Qwen Image 2.1, Flux 2 Pro, Imagen 4, etc. Users can start generating from a text prompt, or upload up to five reference images and describe what needs to be modified (such as changing the background, adjusting colors, or compositing multiple images).
- Multi-model comparison: Each run can request 1 to 4 variants for side-by-side comparison. Each variant is an independent generation task, and only tasks actually started are billed.
- Video generation: The same account can use 21 video models; video generation requires purchased credits.
- Transparent costs: Image generation consumes 4 to 32 credits; a standard Nano Banana image consumes 8 credits, and Nano Banana Pro consumes 32. The generate button shows the exact cost before the operation. If generation fails, credits are automatically refunded, up to five times per day.
- Safety review: Reference images and generation results undergo automatic safety checks; prompts and reference images that fail review are rejected before billing.
- Additional tools: Free practical tools such as a prompt library, image compressor, format converter, and QR code generator are provided. All generated images and videos are saved in "My Creations."
How to Use
- Describe the image: Enter text descriptions such as subject, style, lighting, and composition; you can use the prompt suggestions below the prompt box or open the prompt library for inspiration.
- Select model and format: Nano Banana is selected by default; you can switch models to compare results, and choose the aspect ratio (1:1, 16:9, 9:16, 4:3, or 3:4) and number of variants.
- Generate or edit: Click the generate button. To modify an existing image, switch to edit mode, add up to five reference images, and describe the modifications. Logging in is required for first use, and entered content will be preserved.
- Download and save: Results appear in the outputon the board and available for download at any time. All created images and videos are also saved under "My Creations."
Use Cases
- Quickly generate everyday images and edits
- When images containing text are needed, GPT Image 2.5 or Qwen Image 2.1 can be used
- For stylized scenes, Seedream 5.0 can be used
- For the most refined results, Nano Banana Pro or Imagen 4 Ultra can be used
- Video generation (using purchased credits)
Limitations
- Free credits (sign-up rewards and invitation rewards) apply only to images and other tools and cannot be used for video generation.
- New accounts can receive 16 free credits in supported regions, equivalent to 2 standard images, and the sign-up reward is limited to once per person.
- Commercial use rights may vary by model provider and use case, and the terms of service and provider terms must be consulted.
- Nano AI is an independent service and is not affiliated with Google.
- Video model costs vary by model and duration; for example, a 5-second Kling 2.6 video consumes 55 credits.
- Currently, detailed information such as subscription prices, the list of supported regions, and maximum generation resolution is not provided.
