A lower-cost variant of Krea 2 Large, trading some fidelity for faster and cheaper generation while keeping the same photorealistic style focus.
Test this model instantly in the Console Playground โ no code required
Copy usage instructions for Claude, ChatGPT, or other AI
curl -X POST "https://api.core.today/v1/predictions" \
-H "Content-Type: application/json" \
-H "X-API-Key: cdt_your_api_key" \
-d '{
"model": "krea/krea-2-medium",
"input": {
"prompt": "A ceramic coffee mug on a wooden table near a window, soft morning light, shallow depth of field",
"aspect_ratio": "4:3",
"creativity": "medium"
}
}'| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
prompt | string | Yes | - | Text prompt describing the image. |
aspect_ratio | string | No | 1:1 | Aspect ratio of the generated image. 1:14:33:216:92.35:14:52:39:16 |
creativity | string | No | medium | How far the model expands on your prompt. 'raw' renders only what you describe; 'low' fills in obvious gaps; 'medium' adds reasonable interpretation; 'high' takes meaningful creative liberty. rawlowmediumhigh |
style_reference_images | array | No | [] | Up to 10 reference images whose style should be transferred to the output. Their style is extracted and applied with the strength controlled by `style_reference_strength`. |
style_reference_strength | number | No | 0.5 | How strongly to apply the style of the reference images. 0 means no influence, 1 means the style dominates. Applied to all style reference images. |
moodboard_id | string | No | - | Optional Krea moodboard UUID. Moodboards must first be created in the Krea webapp (https://www.krea.ai/moodboards). Get the UUID from the moodboard share URL: `?share=<uuid>`. |
moodboard_strength | number | No | 0.35 | How strongly the moodboard shapes the output. |
seed | integer | No | - | Random seed. Set for reproducible generation. |
There are 3 ways to provide files for the style_reference_images parameter:
Attach files directly to POST /v1/predictions/upload. No separate upload step needed.
curl -X POST "https://api.core.today/v1/predictions/upload" \
-H "X-API-Key: cdt_your_api_key" \
-F "model=krea/krea-2-medium" \
-F 'input={"prompt":"your prompt here"}' \
-F "file:style_reference_images=@your_file.png"Pass a publicly accessible URL directly. You can also use file_url from the Storage API.
{
"model": "krea/krea-2-medium",
"input": {
"prompt": "your prompt here",
"style_reference_images": ["https://example.com/image.jpg"]
}
}Common parameters used when calling POST /v1/predictions.
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
model | string | Yes | - | Model identifier |
input | object | Yes | - | Object containing the model-specific parameters from the table above |
output_folder | string | No | - | Folder path for output files (max 256 chars, '..' not allowed) |
webhook_url | string | No | - | Webhook URL to call on completion |
is_public | boolean | No | false | If true, output files are also available via permanent public URLs |
A quick photorealistic draft using the default creativity level and a 4:3 aspect ratio.
curl -X POST "https://api.core.today/v1/predictions" \
-H "Content-Type: application/json" \
-H "X-API-Key: cdt_your_api_key" \
-d '{
"model": "krea/krea-2-medium",
"input": {
"prompt": "A ceramic coffee mug on a wooden table near a window, soft morning light, shallow depth of field",
"aspect_ratio": "4:3",
"creativity": "medium"
}
}'Transfers the look of a style reference image onto a new scene at moderate strength.
curl -X POST "https://api.core.today/v1/predictions" \
-H "Content-Type: application/json" \
-H "X-API-Key: cdt_your_api_key" \
-d '{
"model": "krea/krea-2-medium",
"input": {
"prompt": "A city street at dusk with neon signs reflecting on wet pavement",
"aspect_ratio": "16:9",
"style_reference_images": [
"https://example.com/reference-style.jpg"
],
"style_reference_strength": 0.5
}
}'POST /v1/predictions