Events

Current and upcoming events

View all
No events available yet.

Model prompt collection

Gemini AI Photo Editing Prompts

Choose a photo idea, open its full prompt, then adapt the changes for your own image in Gemini.

How to adapt a photo editing prompt for Gemini

  • Open an example above and copy its original prompt. Choose which parts describe your intended edit, such as a new background or outfit.
  • Upload the photo you want to edit in Gemini. Explain what to change and what to preserve, including the subject, pose and framing.
  • Compare the result with your original photo. If a detail drifted, request a smaller edit and repeat the details that need to stay the same.

Example to adapt

Edit my uploaded portrait: replace the background with a softly blurred seaside cafe at sunset. Keep the person, clothing, pose and crop unchanged. Match the new background lighting to the subject. Do not add people or text.

These prompts are starting points to adapt. Inclusion here does not verify which tool created an example image or mean that its prompt has been tested in this tool. Results can differ from the examples.

Using these prompts with Nano Banana in Gemini

Nano Banana is Google's name for its image-generation model family available in Gemini. Use the same prompt idea from this library: describe the subject, scene and style, or upload a photo and explain the edit. You do not need a separate copy of the prompt for each tool. Model options and access can change; see the official Gemini guidance below.

The existing example images in this collection were generated with ChatGPT. They are visual references, not verified Nano Banana results. Compare your own output before assuming a prompt works the same way in both tools.

Official Gemini image guidance

Common questions

Will the same prompt give the same image in ChatGPT and Nano Banana?

You can reuse the main description, but do not expect identical results or a fixed similarity percentage. Adjust wording for the tool and task, especially when a prompt refers to an uploaded photo. Keep the original prompt and save any useful changes separately.

Can I use a prompt without uploading a photo?

Yes, ask Gemini to create a new image from a description. Remove instructions referring to an uploaded photo and describe the subject instead. Upload your photo when the task is to edit that existing image.

Will a photo edit keep every detail unchanged?

Do not assume it will. Compare the face, pose, clothing and background with the original before using the result. Gemini image features also depend on account, age and regional availability; check Google's guidance for current access.

Explore More

Explore more categories

Tag Cloud

Popular tag cloud

Explore AI image prompt tags for ChatGPT, Gemini, portraits, couples, cinematic scenes, fantasy photos, birthdays, fashion, and more creative styles.