Qwen-Image-2.1-Turbo
An image generation and editing model with an eight-step sampling schedule, downloadable weights and hosted API access.
Gemini Nano Banana 2.1
ai.google.dev
Generate and conversationally edit images with reference images, search grounding and output resolutions up to 4K.
Visit website
Pricing
Paid
Available on
Google AI Studio, Gemini API
Category
Ai Image Tools
What is Gemini Nano Banana 2.1?
Gemini Nano Banana 2.1 is Google's stable image generation and conversational editing model. It accepts text, images, video and PDFs and returns images and text through the identifier gemini-nano-banana-2.1.
Designers, marketing teams and developers can use it for infographics, marketing assets and iterative edits. Video references can inform thumbnails, posters or artwork.
Key features
Conversational editing
Refine an image through successive instructions, retaining previous interactions as context.
Resolution and panorama controls
Choose 1K, 2K or 4K output and panoramic ratios including 1:4, 4:1, 1:8 and 8:1. The default resolution is 1K.
Multiple reference images
Combine up to 14 references, with support for consistency across four characters and fidelity across ten objects.
Text within images
Generate stylized text for menus, diagrams, infographics and marketing materials.
Search grounding
Use Google Web and Image Search as context for image generation.
Who is it for?
Marketing teams developing campaign visuals
Explore asset variations using product references and successive editing instructions.
Designers preparing wide compositions
Create panoramic concepts for banners and other wide placements.
Educators creating visual explanations
Draft diagrams and infographics with integrated labels.
How to use Gemini Nano Banana 2.1
- Open the model's Google AI Studio entry or prepare a Gemini API request.
- Select gemini-nano-banana-2.1 and provide a description with relevant references.
- Choose the output resolution and aspect ratio.
- Generate an image, then submit follow-up instructions to refine it.
Gemini Nano Banana 2.1 pricing
Paid
The API meters input and output tokens. Batch token rates are half standard rates. Web and Image Search grounding includes a monthly allowance shared across Gemini 3 and newer models; additional usage is charged separately.
Standard API
USD 1.50 per million input tokens; USD 7.50 per million text and thinking output tokens; USD 30 per million image output tokens
Metered token usage; no recurring subscription interval
Paid-tier Gemini Nano Banana 2.1 rates. Input and output charges can both apply.
Batch API
USD 0.75 per million input tokens; USD 3.75 per million text and thinking output tokens; USD 15 per million image output tokens
Metered batch token usage; no recurring subscription interval
Paid-tier Gemini Nano Banana 2.1 batch rates.
What to know before you use it
- There is no free-tier API allocation for this model. Search grounding charges apply after a shared monthly allowance.
- Audio input, function calling, structured outputs, caching and Live API are unsupported. The model also lacks 512px output.
- Search grounding cannot use real-world images of people. Generated images include SynthID, and requested image counts may not be followed exactly.
Frequently asked questions
Can it generate video?
Video is supported as reference input for image creation. Outputs are images and text.
Are image-output prices the full request cost?
No. Input tokens, text and thinking output, image output and grounding beyond the shared allowance are separate billing components.
Gemini Nano Banana 2.1 alternatives
Gemini 3 Pro Image
Google positions Nano Banana Pro for complex professional asset production and precision creative control.