Nano Banana turns words or photos into the image you had in mind. Generate from a prompt or edit and combine up to 14 reference photos, output at 1K, 2K, or 4K, and see the credit cost before you generate.
Pass up to 14 reference photos to blend, transfer style, and edit.
Test at 1K, finish at 4K. Pick the resolution that fits the job.
Examples made on Kavel with this model.
Generate from text, or edit and combine from photos. These are real output examples, from a single generated subject to a multi-photo composite. Each example below started from a plain prompt or a handful of reference photos, so you can judge the kind of result to expect before you spend a single credit.
Nano Banana is Google's Gemini image model for generating and editing images. On top of text-to-image, it takes up to 14 reference photos so you can edit, combine, and transfer style in a single run, which is where it pulls ahead of a plain prompt box. Polishing one clean shot and merging several photos into one composition are both in scope, and prompts can run up to 20,000 characters when a scene needs detail.
Output comes at 1K, 2K, or 4K in PNG or JPEG. On Kavel, Nano Banana runs, the credit estimate updates before you commit, and failed generations are refunded automatically (unless it broke the content policy). Because generation, editing, and compositing share one screen and one credit meter, a job that would normally cross a generator, an editor, and an upscaler stays in one place from first draft to final frame.
Creative engine
Choose an input method, set resolution, aspect ratio, and photo count, and the credit estimate updates before you generate.
Text-to-image generates a picture from a written description alone, with prompts up to 20,000 characters for detailed scenes.
Image-to-image edits and combines up to 14 reference photos, so composites and style transfer happen in one run.
Output at 1K, 2K, or 4K lets you draft cheaply and then finish at the resolution the job needs.
Reach for it when you want more than one clean shot: when several photos need to become one composition, and when references should steer the result. A lighter Lite variant keeps quick drafts cheap while the full model handles the final, high-resolution frame. In practice that means fewer tools to learn, one bill to watch, and a faster path from a rough idea to a frame you can actually ship, whether you are polishing a single portrait or assembling a product scene from several source photos.
Up to 14 reference photos means composites and style transfer that a single image cannot do finish in one generation.
Draft with the lighter Lite variant (a flat 8 credits at any resolution), then switch to the full model when you need several reference photos or 2K and 4K detail.
The estimate for your resolution and photo count shows before you generate, and failed generations cost nothing.
From text generation to 14-photo composites, the model covers roles that usually take several separate tools, kept in one place with one credit meter.
Describe the scene you want in up to 20,000 characters and generate it.
Pass several photos to edit and combine. JPEG, PNG, and WebP are accepted.
Bring elements and looks from several images together into one frame.
Output 1:1, 16:9, 9:16, 4:3, 3:4, 21:9, and more to fit the use.
The generation parameters for this model, from the provider's published spec.
Four steps from prompt to finished image.
Choose text-to-image, or image-to-image from photos.
Write the image you want, or upload up to 14 reference photos.
Choose resolution, aspect ratio, and photo count.
Confirm the estimate, run it, and download the image.
From a single social graphic to product visuals, here is where the model earns its place.
Turn out post and thumbnail visuals quickly, in the ratio each platform wants.
Combine reference photos to tidy up the visuals around a product.
Merge several photos and swap style or elements to reach a final frame, without bouncing between a separate editor and generator.
Turn an idea into a finished picture fast to check a direction before you commit time to it, then re-run at a higher resolution once the look is right.
One credit pool covers Nano Banana images and Seedance and Veo video. The cost shows before every run, and failed jobs are not charged. Use a subscription for ongoing work, or a one-time pack when you just need to top up.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
One-time top-ups — buy extra credits any time you run low.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
Charges appear as “KAVEL AI” on your card statement. You can cancel any time from Settings → Billing; cancellation takes effect at the end of the period you already paid for.
Powered by
Use one balance across every supported image and video model — eleven of them today, listed below — and check the credit cost before each request.
Video models
Image models
Model credit guide
You're only charged for successful generations. The exact estimate in the generator varies by model, length, resolution, audio, and number of images.
| Type | Model | Credit cost |
|---|---|---|
| Video | Seedance 2.5 | The long-take tier, priced per second: 5s 480p ≈ 211 credits, 5s 720p ≈ 473. A full 30s take runs ≈ 1,260 at 480p and ≈ 2,835 at 720p. Supplying a reference clip lowers the per-second rate. |
| Video | Seedance 2.0 | 5s 720p image-to-video ≈ 188 credits; text-to-video ≈ 308 credits. Scales with resolution and length. |
| Video | Seedance 2 Fast | Faster and lower cost. 5s 720p text-to-video ≈ 248 credits. |
| Video | Seedance 2 Mini | The cheapest Seedance tier. 5s 720p ≈ 154 credits, 5s 480p ≈ 72 credits. |
| Video | Seedance 1.5 Pro | Audio doubles the rate. 5s 720p ≈ 27 credits silent, ≈ 53 with audio; 1080p ≈ 57 and ≈ 113. |
| Video | Veo 3.1 | Billed per video, not per second (Lite tier). About 45 credits at 720p and 53 at 1080p. |
| Video | Kling 3.0 | Audio raises the rate. 5s 720p ≈ 105 credits silent, ≈ 150 with audio; 1080p ≈ 135 and ≈ 203. |
| Video | MiniMax H3 | Fixed 2K, no resolution ladder. Priced per second — a 5s clip ≈ 158 credits. |
| Image | Nano Banana 2 | Generate or edit from text and images. About 8 credits per 1K image, 12 at 2K, 18 at 4K. |
| Image | Nano Banana Pro | Consistent run times across generations. About 12 credits per 1K or 2K image, 21 at 4K. |
| Image | Nano Banana 2 Lite | Faster, simpler variant. A flat 8 credits per image at every resolution. Start here to test. |
| Image | GPT Image 2 | The lowest-cost image model here. About 3 credits per 1K image, 6 at 2K, 12 at 4K. |
| Image | Seedream 5.0 Lite | Flat pricing — about 8 credits per image whether you generate or edit. |
Common questions about the Nano Banana AI image generator.
It is Google's Gemini image model. It generates images from text and can take up to 14 reference photos to edit and combine. Prompts run up to 20,000 characters, and output comes at 1K, 2K, or 4K in PNG or JPEG.
Credits depend on resolution and photo count. A single 1K image is about 8 credits, for example. The estimate always shows before you generate, and failed generations cost nothing.
Lite is the lightweight tier. It takes one source image and costs a flat 8 credits at any resolution, while the full model accepts up to 14 reference photos and rises to 12 credits at 2K and 18 at 4K. At 1K the two cost the same, so Lite is the simpler and faster option rather than the cheaper one.
Yes. Upload up to 14 reference photos and it can edit, combine, and transfer style from them. JPEG, PNG, and WebP are accepted, up to 30MB each.
Describe the subject, style, composition, and lighting in detail. Passing reference photos makes it easier to match color and framing.
Check the use of any generated image against your own use case and license terms. If you pass reference photos, respect the rights to those photos as well.
Open the generator, pass text or a photo, check the credits, and generate.
See the credits for your resolution and photo count before you generate.
