Generating an image
To query an image model, use the.images
method and specify the image model you want to use.

Provide reference image
Use a reference image to guide the generation:
Supported Models
See our models page for supported image models.Parameters
Parameter | Type | Description | Default |
---|---|---|---|
prompt | string | Text description of the image to generate | Required |
model | string | Model identifier | Required |
width | integer | Image width in pixels | 1024 |
height | integer | Image height in pixels | 1024 |
n | integer | Number of images to generate (1-4) | 1 |
steps | integer | Diffusion steps (higher = better quality, slower) | 1-50 |
seed | integer | Random seed for reproducibility | any |
negative_prompt | string | What to avoid in generation | - |
frame_images | array | Required for Kling model. Array of images to guide video generation, like keyframes. | - |
prompt
is required for all models except Klingwidth
andheight
will rely on defaults unless otherwise specified - options for dimensions differ by model- Flux Schnell and Kontext [Pro/Max/Dev] models use the
aspect_ratio
parameter to set the output image size whereas Flux.1 Pro, Flux 1.1 Pro, and Flux.1 Dev usewidth
andheight
parameters.
Generating Multiple Variations
Generate multiple variations of the same prompt to choose from:
Custom Dimensions & Aspect Ratios
Different aspect ratios for different use cases:
Quality Control with Steps
Compare different step counts for quality vs. speed:
Safety Checker
We have a built in safety checker that detects NSFW words but you can disable it by passing indisable_safety_checker=True
. This works for every model except Flux Schnell Free and Flux Pro. If the safety checker is triggered and not disabled, it will return a 422 Unprocessable Entity
.
Troubleshooting
Image doesn’t match prompt well- Make prompt more descriptive and specific
- Add style references (e.g., “National Geographic style”)
- Use negative prompts to exclude unwanted elements
- Try increasing steps to 30-40
- Increase
steps
to 30-40 for production - Add quality modifiers: “highly detailed”, “8k”, “professional”
- Use negative prompt: “blurry, low quality, distorted, pixelated”
- Try a higher-tier model
- Use
seed
parameter for reproducibility - Keep the same seed when testing variations
- Generate multiple variations with
n
parameter
- Specify
width
andheight
explicitly - Common ratios:
- Square: 1024x1024
- Landscape: 1344x768
- Portrait: 768x1344
- Ensure dimensions are multiples of 8