Skip to main content
POST

Authorizations

​
x-magnific-api-key
string
header
required

Your Magnific API key. Required for authentication. Learn how to obtain an API key

Body

application/json
​
prompt
string
required

Text description of the image you want to generate.

FLUX.2 [flex] offers flexible, professional-grade image generation with fine-grained control over guidance, steps, and a wide resolution range for large-format output.

Tips for better results:

  • Be specific about subjects, scenes, and visual details
  • Describe lighting, atmosphere, and mood
  • Mention art style or photographic techniques if desired

Examples:

  • Simple: "a cat sitting on a windowsill"
  • Detailed: "a fluffy orange tabby cat sitting on a rustic wooden windowsill, golden hour lighting, soft focus background"
  • Creative: "futuristic cityscape at night with neon lights reflecting on wet streets, cyberpunk aesthetic, dramatic lighting"
Example:

"a beautiful sunset over the ocean with dramatic clouds"

​
width
integer
default:1024

Width of the image in pixels.

Valid range: 256 to 1920 pixels Default: 1024 pixels

Common sizes:

  • Square: 1024×1024
  • Landscape: 1920×1080 (16:9)
  • Portrait: 1080×1920 (9:16)
Required range: 256 <= x <= 1920
Example:

1024

​
height
integer
default:768

Height of the image in pixels.

Valid range: 256 to 1920 pixels Default: 768 pixels

Tip: Total megapixels affect generation cost. Higher resolution = higher cost.

Required range: 256 <= x <= 1920
Example:

768

​
seed
integer | null

Random seed for reproducible results.

Use case: Generate variations of the same image by using the same seed with slightly modified prompts.

Valid range: 0 to 4,294,967,295 Default: Random seed (if not provided)

Required range: 0 <= x <= 4294967295
Example:

42

​
guidance
number | null
default:3

Guidance scale for the generation. Higher values make the model follow the prompt more closely. Valid range: 1.5 to 10.0.

Required range: 1.5 <= x <= 10
​
steps
integer | null
default:50

Number of inference steps. More steps generally produce higher quality but take longer. Valid range: 1 to 50.

Required range: 1 <= x <= 50
​
safety_tolerance
integer | null
default:5

Tolerance level for input and output moderation. Between 0 and 5, 0 being most strict, 5 being least strict.

Required range: 0 <= x <= 5
​
prompt_upsampling
boolean
default:false

Automatically enhance and expand your prompt for better generation results.

When to enable:

  • Using simple, short prompts
  • Want AI to add creative details
  • Experimenting with variations

When to disable:

  • Using detailed, specific prompts
  • Need exact control over output
  • Following brand guidelines
Example:

false

​
output_format
enum<string> | null

Format of the output image

Available options:
jpeg,
png
​
input_image
string | null

Base64-encoded input image for image-to-image generation or image editing.

Supported formats: JPEG, PNG, WebP Use cases:

  • Modify existing images
  • Style transfer
  • Composition refinement

Note: Can provide up to 4 input images using input_image, input_image_2, input_image_3, and input_image_4.

​
input_image_2
string | null

Second base64-encoded input image for multi-image generation.

​
input_image_3
string | null

Third base64-encoded input image for multi-image generation.

​
input_image_4
string | null

Fourth base64-encoded input image for multi-image generation.

​
webhook_url
string<uri>

Optional callback URL that will receive asynchronous notifications whenever the task changes status. The payload sent to this URL is the same as the corresponding GET endpoint response, but without the data field.

Example:

"https://www.example.com/webhook"

Response

OK - Task created successfully

​
data
object
required
Example: