Why Nano Banana 2?
Text to image
Describe a scene and get a high-fidelity still.
Conversational edit
Pass up to 14 images and rewrite them with a prompt.
Readable text
Crisp in-image lettering, including multilingual type.
1K, 2K, 4K
Pick resolution per cook. Higher takes longer.
Character lock
Keep the same face across stills with reference images.
Pricing
| Res | Credits |
|---|---|
| 1K | ✦ 10 |
| 2K | ✦ 15 |
| 4K | ✦ 20 |
How Nano Banana 2 works
Nano Banana 2 is Google’s latest image model, built on Gemini 3.1 Flash Image. It is the high-efficiency counterpart to Nano Banana Pro — Pro-level fidelity with Flash-level speed. One cook bar, two capabilities. The number of stills you attach is the only switch.
Outputs are stills at 1K, 2K, or 4K. Attach PNG, JPEG, or WebP. Up to 14 reference images for fusion, style transfer, or character lock.
Capabilities
- text-to-image — 0 stills. The prompt becomes the whole frame.
- image-edit — 1 to 14 stills. Transform, restyle, fuse, or keep a character while you rewrite the scene.
How to use
Everything happens in the cook bar at the bottom of the page.
- 1. Prompt — describe the still, or the edit you want on attached images.
- 2. Stills — skip them for generate, or attach up to 14 for edit / fusion / character reference.
- 3. Settings — ratio, 1K / 2K / 4K, optional web search or image search.
- 4. Cook — the button shows the resolution price. Closing the tab after cook starts is fine; we email you when the still is ready.
Text to image
Generate a still from a written description — photoreal, illustration, product, infographic, or readable in-image type.
- How — leave the image tray empty. Write the prompt, set ratio / resolution, cook.
- Best for original scenes, logos, comics, and product mockups from text alone.
Image edit
Transform existing stills with a prompt. One image is a directed edit. Several images fuse style, character, and layout into one frame.
- How — attach 1–14 PNG, JPEG, or WebP stills. Describe the change, then cook. Use match_input_image to keep the source aspect.
- Best for restyles, background swaps, multi-image combination, and character-consistent series.
Settings
-
prompt - Required. Describe the image to generate, or the edit to apply.
-
images - Optional. Skip for text-to-image. Attach 1–14 stills (PNG, JPEG, WebP) to edit, fuse, or use as character/style reference.
-
aspect_ratio - 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9, plus ultra-wide 1:4, 4:1, 1:8, 8:1. match_input_image copies the first attached still.
-
resolution - 1K (default), 2K, or 4K. Higher resolutions take longer.
-
google_search - Optional. Ground the image in live web search (weather, scores, recent events).
-
image_search - Optional. Pull web images as visual context. Also turns on web search.
About
Google announced Nano Banana 2 on February 26, 2026 as their best image generation and editing model — combining Pro-level visual quality with Flash-level speed. It is served here via Replicate as google/nano-banana-2.
FAQ
- What can Nano Banana 2 generate on Soup.video?
- Text-to-image when the tray is empty. Image-edit when you attach one or more stills — style transfer, background swaps, multi-image fusion, and character-consistent edits. Up to 14 reference images.
- How do I use Nano Banana 2?
- Write a prompt. Attach stills or skip them. Set ratio and 1K / 2K / 4K. Optionally turn on web or image search. Cook. Closing the tab after cook starts is fine; Soup.video emails you when the still is ready.
- How does text-to-image work?
- Leave the image tray empty. The prompt becomes the whole frame. Set ratio and resolution, then cook.
- How does image-edit work?
- Attach 1–14 PNG, JPEG, or WebP stills. The prompt describes the change — restyle, combine, swap a background, keep a character. Use match_input_image to keep the source framing.
- What resolutions and prices does Nano Banana 2 support?
- 1K is ✦ 10 per image, 2K is ✦ 15, 4K is ✦ 20. Same price for generate and edit.