Create and edit images with Nano Banana 2
Fast multimodal creation with reference-aware edits and readable text
Turn simple prompts and reference images into expressive visuals with Nano Banana 2, also known as Gemini 3.1 Flash Image, using fast iteration, flexible framing, and model-aware editing controls in CinePixel.
Also known as Gemini 3.1 Flash Image
- Create polished visuals from concise natural-language prompts
- Upload a photo and reimagine its look or composition
- Render useful text and structured details inside images
- Explore concepts quickly across multiple aspect ratios
Create from text
Create finished images, concept art, product shots, and visual drafts from text.
Nano Banana 2
Create from text
Create finished images, concept art, product shots, and visual drafts from text.
Create from text
Create finished images, concept art, product shots, and visual drafts from text.
Compare matched model alternatives
These catalog-backed models share the strongest verified capabilities with Nano Banana 2. Review the evidence before choosing a workspace.
Explore image generators by task
Jump into a focused image generator for a specific creative task, then continue in its canonical CinePixel workspace.
Move from an idea to a useful visual
Load a model-aware starting point, adjust it in the generator, and keep full control over when creation begins.
Fast iteration with visual continuity
Nano Banana 2 supports creators who need quick variations, reference-aware changes, and dependable subject consistency.
- Text to Image
- Credits per generation
- 12-20 creditsPricing source updated
- Supported aspect ratios
- 16:9 · 9:16 · 1:1 · 4:3 · 3:4 · 3:2 · 2:3
- Output format
- PNG · JPG
- Reference image capacity
- Up to 14 reference images
- Resolution
- 1K / 2K / 4K
- Image to Image
- Credits per generation
- 5-24 creditsPricing source updated
- Supported aspect ratios
- 16:9 · 9:16 · 1:1 · 4:3 · 3:4 · 3:2 · 2:3
- Output format
- PNG · JPG
- Reference image capacity
- Up to 14 reference images
- Resolution
- 1K / 2K / 4K
Exact options are revalidated in the workspace for the selected workflow and aspect ratio.
- Rapid concept exploration from concise natural-language prompts
- Reference-led transformations with recognizable subject continuity
- Useful embedded text for campaign and social compositions
- High-resolution outputs across common publishing aspect ratios
How to use Nano Banana 2
Use the live workbench in four deliberate steps, then review the result before you decide what to create next.
Choose a model and workflow
Start from this canonical model route and select the workflow that matches your brief, such as Text to Image or Image to Image when available.
Describe the visual direction
Write the subject, composition, lighting, materials, and must-keep constraints in the prompt, putting the most important details first.
Add references and adjust settings
Add reference images when the selected workflow supports them, then adjust only the ratio, resolution, format, and other controls shown for that workflow.
Review before creating
Inspect the prompt, references, model, and output settings, then choose Generate when the brief is ready.
Nano Banana 2 questions
When should I use Image to Image?
Use it when a subject, product, character, or composition from an existing image must guide the new result.
How do I keep a subject consistent?
Use a clear reference and state exactly which identity, clothing, pose, or composition details must remain unchanged.
Will an example overwrite my route settings?
No. It updates only the prompt text and keeps the current model route, mode, references, and generator settings.
Ready to shape the next image?
Return to the live workbench to adjust your prompt, inputs, and output settings before you create.
Compare image models
Choose a CinePixel image model by workflow, provider, and current credit range, then continue in its canonical workspace.