Create & edit images

9brains can create and edit images directly in chat. Just describe what you want to see, and the AI automatically picks the right tool and model.

Generate images

Describe the desired image in your message. The AI automatically detects that an image should be created and uses the image generation tool.

Example requests:

• “Create a photo of a modern office with wooden furniture and plants.”
• “Generate an illustration on the topic of teamwork in watercolor style.”
• “Make a product photo of a white coffee cup on a wooden table with soft light.”

Tips for better results:

• Describe style, colors and composition as precisely as possible
• Name the desired image type (photo, illustration, icon, graphic, …)
• Provide context (e.g. “for a presentation” or “in cartoon style”)

The generated image appears directly in the chat. To save it, hover over the image and click the download icon in the top right corner.

Image format and resolution

You can state the aspect ratio and the size directly in your message. Without one, the AI picks a suitable format.

Example requests:

• “Create a banner at 3840x990 pixels.”
• “Make the image in portrait 9:16 for Instagram Stories.”
• “Create a product photo in 4:3.”

Not every model can do every format. Nano Banana Pro and Nano Banana 2 handle all common aspect ratios from square to very wide banners (21:9) and deliver a high resolution by default. GPT Image 2 accepts arbitrary pixel dimensions but stays below 3:1. FLUX gives you free formats when creating an image, but only squares when editing one.

If you ask for a format no model produces directly (a very wide advertising banner, for instance), the image is created in the closest available format and then cropped to your dimensions. The AI tells you how much was cropped away, so you can check that nothing important is missing.

Edit images

You can have existing images edited in chat. Upload an image and describe the desired change.

How it works:

1. Upload an image in chat (or use a previously generated image)
2. Describe the desired change

Example requests:

• “Make the background of this image blue.”
• “Remove the text from the image.”
• “Change the color of the logo from red to green.”

The original is preserved, the edited version is saved as a new file. The images in a chat are numbered consecutively: bild-1.png, bild-2.png and so on. Very large images are stored as .jpg so they can be used for further editing.

Supported formats: PNG, JPEG, WebP (max. 50 MB per file)

Combine images (multi-image)

You can have elements from one image inserted into another, ideal for compositing and product visualizations.

How it works:

1. Upload the main image (e.g. an empty living room)
2. Upload one or more reference images (e.g. a piece of furniture)
3. Describe what the AI should do

Example requests:

• “Place the couch from the second image in the right corner of the living room.”
• “Insert the product from the reference image onto the table in the main image.”
• “Place the logo on the T-shirt in the photo.”

Note: For multi-image, a compatible model is automatically selected. You don’t need to worry about it.

Available image models

9brains supports several specialized image models from different providers:

Model	Location	GDPR-compliant?	Rating (create / edit)
Google
Nano Banana Pro	USA (DPF certified)	Yes	1102 / 1099
Nano Banana 2.1	USA (DPF certified)	Yes	— / —
Nano Banana 2	EU	Yes	1125 / 1108
Black Forest Labs
FLUX.2 Pro	Paris, France	Yes	1004 / —
FLUX.2 Flex	Global (DPF certified)	Yes	1028 / 1006
OpenAI
GPT Image 2	Global (DPF certified)	Yes	1172 / 1122
GPT Image 2.5 Flare	Global (DPF certified)	Yes	1191 / 1162
GPT Image 2.5 Sunburst	Global (DPF certified)	Yes	1197 / 1182

Rating: Elo score from blind comparisons, in which people pick the better of two images for the same request without knowing the models. Higher is better, a dash means not rated or, for editing, not supported. Source: Artificial Analysis

The AI uses your workspace’s image model, a separate one each for creating and for editing. In EU mode these are automatically models that run in the EU. Under Settings → AI models you can also set personal image models that then apply in all your chats. You can also request a specific model, e.g.: “Create an image with FLUX.2 Pro: …”

Which models are available is determined by your administrator (see Groups & AI models).

Frequently asked questions

Can I switch the image model?

Yes. You can request a specific model in your message. Otherwise the AI uses your personal image model or, if you have not set one, your workspace’s. In EU mode, only models that run in the EU are available.

What happens to my generated images?

All generated and edited images are stored in your workspace and can be retrieved at any time via the chat history. You can download them at any point using the download icon in the top right corner of the image.

Which image formats can I upload?

PNG, JPEG and WebP, up to 50 MB per file.

Is my original image preserved?

Yes. Image editing always creates a new file. The original is not modified.

My image was rejected by a content check. What now?

Every provider screens requests automatically for unwanted content. That check occasionally fires on harmless subjects too, especially on photos of people. It cannot be switched off, it sits with the provider of the model.

The way around it is a model from a different provider, because the checks differ in strictness. Just say in your message which model to use, for example: “Try it with Nano Banana Pro.” For photos of people, Nano Banana Pro is the most reliable choice.

Can I also have videos generated?

Currently 9brains supports text and image generation. Video generation is not part of the feature set at this time.