This article is also available in:

How to use Gen AI Effect feature to create custom image effects

You can preset an AI prompt so that your customer’s uploaded photo is automatically transformed into a specific style or effect. For example, you could turn their photo into a Ghibli-style or Disney-style artwork. This lets you offer fun, AI-generated product designs without any extra steps from the customer.

Please continue reading this Article to learn more about this feature!

Connecting Image Generation Service

To use this feature, you’ll need an API key for your chosen AI model. Usage will be charged based on the model you select. Please refer to the following article to connect to an AI model.

Inside Teeinblue Portal, navigate to Extras > 3rd-party Services > Gen Ai Effects
Select the Model you want to use, for example, GPT Image.

Insert your own API Key here

Creating the Artwork with AI Image Prompt feature

Go to Artwork editor > edit an image layer

In Personalization Options, choose Upload photo. You can read more about that feature here.

Under the Image effects section, select Gen AI Effect and your desired AI Model.

You can choose "Default settings" to follow the default service you selected in Extra Services.

You can assign a different AI model to each Upload Photo Layer in your artwork. This gives you flexibility to use one model for a specific Upload Photo Layer and another model for a different layer. Since each model may produce different results, you can freely choose which one works best.

Make sure you’ve connected the APIs for all the models you plan to use.

Be sure to enter a detailed prompt — the more specific it is, the better the result will be. You can also click the Suggested Prompt button to see our pre-made prompts for popular styles.

NOTE: if your prompt contains words or subjects that violate the terms of use of the selected AI model, the app will simply return the original uploaded image instead. You can view their policies here.

Since generating images with the AI model may take some time, we recommend adding a note in the Add instruction for customer box to let customers know.

If you want the Gen AI effect to generate an image with a transparent background, make sure to turn on the Remove background (transparent) option.

Creating the Campaign

Once you're happy with your settings, be sure to launch it to your Storefront!

Add a reference image for Gen AI Effects

The Reference Image feature lets you maintain a consistent style or outfit across all generated results. By uploading a reference image in the Artwork Editor, you can guide the Gen AI Effect to match that look. For example, you can turn every customer photo into the same “kimono” style. This helps create more stable and cohesive outputs for your artwork, no matter which AI provider you use (OpenAI, Flux, or Gemini).

For example, both of these uploaded images will be generated in the same Kimono style.

To use this feature, simply navigate to your Artwork Editor, and toggle the Use Reference Image button, then upload your reference image there:

The image uploaded by the Customer (base image or source image) is treated as Image 1, and the reference image is Image 2.

In your prompt, clearly describe what Image 1 is used for, such as what the AI should focus on or keep from it, and what Image 2 provides, like the style, outfit, or background.

Then, write your main prompt below to describe the desired result.

To get the best results, you may need to adjust your prompt so the AI understands and applies the reference image properly. Since output consistency can vary depending on the artwork and provider, results might not be 100% accurate on the first try. Test and fine-tune your prompt to achieve the desired style.

Rate Limit for Photo Uploads

To prevent abuse and avoid unnecessary consumption of third-party AI credits, we’ve introduced a rate-limiting feature for photo uploads. This ensures fair usage while helping you control costs.

You can set a rate limit by going into Teeinblue > Extras > 3d-party Service > Upload limit settings

When a buyer reaches the upload limit, they’ll see a message directly in the Product Page so they know why they can’t continue. You can customize this message in Product page content > Upload photo > Error message when reaching upload limit.

It will look like this for the Customers:

When a buyer is close to the limit (less than 3 uploads left), they’ll see a warning message in Product Page. You can customize this warning in Product page content > Upload photo > Warning message when nearing upload limit.

It will look like this for the Customers:

Note: The upload limit does not apply in the Campaign editor, so you can test without restrictions.

**Please keep in mind the following about this feature: **

Applies per buyer (identified by device, not IP)
Applies per Custom Photo Upload option - not the total number of uploads
Only applies to 3rd-party AI effects. Teeinblue’s built-in effects are not affected.
Resets every hour

Controlling Aspect Ratio in Photo Reposition

When using the Photo Reposition mode, you can control whether the AI generates the final image based on the aspect ratio of the buyer's uploaded photo or the aspect ratio of your original artwork layer.

Before proceeding, please refer to the content regarding the relationship between aspect ratio and layers in this Article.

Look for the option: "Base on layer's ratio instead of buyer's photo".

Disabled (Default): The system uses the buyer's uploaded photo's ratio for AI processing. The image inside the reposition frame will retain the customer's original shape.
Enabled: The system uses the Artwork Layer's ratio for AI processing. The output image will be forced to match the aspect ratio of your reposition frame.

Note: When enabling this option, if your layer's ratio is not supported by the selected AI provider, the final image might look distorted or stretched. Please check the supported ratios in the FAQ article linked above.

Update: Flux 2 Kontext

Flux 2 Kontext introduces several important changes compared to Flux 1 Kontext. Please review the following information carefully before using this model.

Pricing Changes

Flux 2 Kontext uses a different pricing calculation compared to Flux 1 Kontext.

Pricing is based on the output image size rather than a fixed aspect ratio.

For the latest and most accurate pricing details, please refer to the official Flux pricing page:

https://bfl.ai/pricing

Supported Image Dimensions & Aspect Ratio

Flux 2 Kontext supports any aspect ratio, as long as it is defined using Width × Height.
The total pixel count (Width × Height) must not exceed 4,000,000 pixels (4MP).
The minimum value for either width or height is 256px.

Important Setup Notes

When setting up Flux 2 Kontext:

Ensure the output width and height match or remain proportional to the original layer’s aspect ratio.
If the output dimensions do not align with the original layer, the generated image may be stretched or distorted.

Tip: To avoid image deformation, always configure the output dimensions to match the original layer’s aspect ratio.

Comparison between AI Models

Please refer to the following article here.

FAQs

Please refer to the following article here.

Updated on: 16/01/2026

Was this article helpful?

Thank you!