How to generate images with Gemini

That is the consumer-friendly route: no code, no API key and no developer setup. The trade-off is that visible options can differ by platform, account type or Google’s latest interface changes.

A simple prompt formula that works

A strong image prompt gives Gemini enough signal about what should be in the image, where it is happening and what it should look like. Google’s Imagen documentation shows prompts built around elements such as subject, context or background and style, and notes that longer prompts let you add more specific details to the image.

Use this formula:

Create/generate/draw + subject + action + setting + style + aspect ratio if needed

Google also encourages specificity. Instead of a broad prompt like “Create an image of a woman in a red dress,” its Gemini page gives a more detailed example: a young woman in a red dress running through a park.

Prompt templates you can copy

Use these as starting points, then swap in your own subject, location and style.

Product image

Create an image of a ceramic coffee mug on a light oak table, morning window light, minimal Scandinavian style, 1:1.

Poster or thumbnail

Generate a cinematic poster of a Vietnamese street food stall at night, warm lantern light, realistic photography, 16:9.

Editorial illustration

Draw a clean editorial illustration of a person writing AI prompts on a laptop, soft blue background, modern flat style, 4:3.

Clear composition

Create a wide-angle photo of a small robot holding a red skateboard on a city sidewalk, sunset light, realistic street photography, 16:9.

The pattern is the same in each example: the clearer you are about subject, context, style and framing, the more useful direction the model has.

How to edit or combine existing images

Gemini Apps can also work from images you provide. Google Help says you can upload an image and ask Gemini to edit it, or use multiple images to create a new image based on them.

When using reference images, split your instruction into two parts:

• Keep: the person, product, main colours, camera angle or layout.
• Change: the background, lighting, clothing, style, aspect ratio or unwanted details.

For example:

Keep the original product and camera angle, but replace the background with a bright modern kitchen, add soft morning light, and make the image suitable for a square ecommerce listing.

That is more useful than asking Gemini to “make it better” or “make it professional,” because it tells the model what should stay the same and what should change.

When to use the Gemini API instead

If you only need to create a few images, brainstorm concepts or edit visuals manually, Gemini Apps is the right starting point for end users.

If you need image generation inside a product, website, internal tool or automated workflow, use the developer route. Google provides documentation for generating images with Imagen models through the Gemini API.

Google’s Imagen API documentation includes examples of short prompts, longer prompts and aspect-ratio options such as 1:1, 4:3, 16:9 and 9:16 in an API context. Do not assume that every API option will appear exactly the same way in Gemini Apps, or that every Gemini Apps interface feature maps directly to an API setting.

Limits and requirements to know

A few conditions can affect whether you see or can use Gemini image generation:

• Country and language: Google says AI image generation is available in the languages and countries where the Gemini app is available.
• Age and account type: Google Help says users under 18 who are signed in with a personal Google Account cannot use image generation; work or school accounts may have separate restrictions.
• Watermarks: Google says images generated by Gemini include an invisible SynthID watermark and a visible watermark to identify AI-generated content.
• Usage policies: When you generate images with Gemini Apps, Google says you are subject to its Terms of Service, including the Prohibited Use Policy; requests that violate policy may be blocked or handled under those rules.

What if you do not see the image button?

First, check both official entry points: Google Help refers to Create image, while Google’s Gemini image-generation page describes Tools > 🍌 Create images.

If the option still does not appear, the reason may be your region or language, your age, whether you are using a personal Google Account or a work or school account, or restrictions set by the organisation that manages the account.

Bottom line

The short version is: go to gemini.google.com, choose Create image or Tools > 🍌 Create images, write a specific prompt, then ask Gemini to revise the result if needed.

For prompts, focus on three pillars: subject, setting and style. For software products or automation, start with the Gemini API and Imagen documentation rather than relying only on the Gemini Apps interface.