I connected Freepik to my automation yesterday. As far as I understand, Freepik uses its weakest model by default, although according to the information on their website, more powerful models are available via the API.
Unfortunately, in Freepik’s node on Make, there is no option to select a model. (For example, in Fal.ai, this can be done directly in the node.)
Could you please suggest a solution? How can I change the generation model to, say, flux.dev or another one?
NicoN
March 16, 2025, 10:45am
2
Hi! You need to work with custom HTTP request but they can be tricky.
Let’s try this:
Add an HTTP module to your Make scenario
Configure it to make a direct API call to Freepik’s specific model endpoints
For the flux-dev model, set up your HTTP request like this:
URL: https://api.freepik.com/v1/ai/text-to-image/flux-dev
Method: POST
Headers:
Content-Type: application/json
x-freepik-api-key: [your Freepik API key]
Body (JSON):
{
"prompt": "your text prompt here",
"webhook_url": "your webhook URL",
"aspect_ratio": "square_1_1"
}
You can also add styling parameters as needed:
"styling": {
"effects": {
"color": "softhue",
"camera": "portrait",
"lightning": "iridescent"
}
}
For Mystic: https://api.freepik.com/v1/ai/mystic
For Google Imagen 3: https://api.freepik.com/v1/ai/text-to-image/imagen3
For Classic Fast: https://api.freepik.com/v1/ai/text-to-image
The response will include a task ID that you’ll need to use in another HTTP request to get the generated image once the task is complete.
For the second request, you’d do something like:
You might even need to add a delay or a loop that keeps checking until the image is ready.
It’s basically like checking the status of your order until it’s ready for pickup.
You can check out this video where I explain the process with the Blackforest API and FLUX.1:
https://www.youtube.com/watch?v=t5YD9Q3Iy3I
1 Like
Thanks a lot. I will be try. Today i make something similiar with Fal.ai.