Images from the model you pick.

• From a description

  Describe the subject, the style and the light. Choose the aspect ratio, the resolution where the model offers one, and the number of images.

• Edit a picture

  Choose Edit on a result or attach a picture, and describe the change. It needs a model that edits images; Auto picks one.

• Regenerate

  Regenerate a result for another take on the same prompt and settings.

• Use a reference

  Attach an image as the reference for a new picture, with a model that takes images as input.

• Animate in Video

  Animate on a result opens Video with the picture as the first frame of a clip.

• Saved and marked

  Every result is saved to your files, marked as AI-generated and can go into a project.

1. 01

   Describe the picture

   Write the prompt, pick an image model or leave it on Auto, and set the aspect ratio and the number of images.

2. 02

   Check the estimate and send

   The estimated cost shows before the run. The images are made as a background job and saved to your files.

3. 03

   Refine and reuse

   Edit a result, regenerate it, animate it in Video, download it or save it to a project.

• Nano Banana 2 (Gemini 3.1 Flash Image)
• Nano Banana Pro (Gemini 3 Pro Image)
• GPT-5.4 Image 2
• Nano Banana (Gemini 2.5 Flash Image)
• Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)
• GPT-5 Image Mini
• GPT-5 Image

On Auto, ewpire picks an image model for the request and moves on to the next one if that model is unavailable; a model you pick yourself is never replaced.

• Nano Banana 2 (Gemini 3.1 Flash Image): 10 images cost 0.81816 credits, at 0.081816 per image.

Counted from the prices in the model catalog, where one credit is $1. The exact charge of each request appears in Usage.

• One request makes up to 4 images, unless the model sets its own number; the Images setting shows the choices.
• Aspect ratios and resolutions differ by model; Settings lists what the chosen model offers.
• A reference image can be up to 20 MB.
• A workspace can have 50 image, video, music and Consensus jobs queued or running at once.

By model

• Nano Banana 2 (Gemini 3.1 Flash Image): up to 4 images per request. Aspect ratios: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9 and 21:9.
• Nano Banana Pro (Gemini 3 Pro Image): up to 4 images per request. Resolutions: 1K, 2K and 4K. Aspect ratios: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9 and 21:9.
• GPT-5.4 Image 2: up to 4 images per request. Aspect ratios: 1:1, 2:3 and 3:2.
• Nano Banana (Gemini 2.5 Flash Image): up to 4 images per request. Aspect ratios: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9 and 21:9.
• Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image): up to 4 images per request. Aspect ratios: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9 and 21:9.
• GPT-5 Image Mini: up to 4 images per request. Aspect ratios: 1:1, 2:3 and 3:2.
• GPT-5 Image: up to 4 images per request. Aspect ratios: 1:1, 2:3 and 3:2.

• It refuses a prompt for a deepfake or face swap of a real person, or for intimate imagery of a real person, before any credit is reserved.
• It does not swap a model you picked. If that model is unavailable, the request ends with an error and nothing is charged.
• It does not make unmarked pictures: every result is labeled AI-generated, and a generated PNG file carries the mark in its metadata.

• Docs
• FAQ on the modes
• Pricing
• Model catalog