Skip to main content
Lumenfall provides access to image generation models from multiple providers. Each model has different capabilities, quality characteristics, and pricing.

​
Discovering models

Browse available models:
  • Model catalog — Full list with capabilities, pricing, and examples
  • Models API — Programmatic access to available models
# List all available models
models = client.models.list()
for model in models.data:
    print(model.id)

​
Using models

Specify a model by its ID in the model parameter: 
response = client.images.generate(
    model="imagen-3",
    prompt="A mountain landscape at sunset",
    size="1024x1024"
)
To use a specific provider, prefix with the provider slug: 
model="fal/nano-banana-pro"

​
Model parameters

Parameters vary by model. Common parameters include:
ParameterDescriptionNotes
sizeImage dimensionse.g., 1024x1024, 1792x1024
nNumber of imagesSome models only support n=1
qualityOutput qualitystandard or hd (select models)
styleVisual stylevivid or natural (DALL-E 3 only)
Not all models support all parameters. Unsupported parameters are ignored. See the Generate images reference for full details.

​
Model behavior

Prompt revision — Some models (like DALL-E 3) automatically expand prompts for better results. The revised prompt is returned in the response as revised_prompt. Size constraints — Models support different dimensions. If you request an unsupported size, the request may fail or use the closest supported size. Rate limits — Some provider/model combinations have lower rate limits. If you hit limits, requests automatically route to alternative providers when available.

​
Pricing

Pricing varies by model, size, and quality. Use dry run mode to estimate costs before generating: 
curl "https://api.lumenfall.ai/openai/v1/images/generations?dryRun=true" \
  -H "Authorization: Bearer $LUMENFALL_API_KEY" \
  -d '{"model": "imagen-3", "prompt": "test", "size": "1024x1024"}'
See Billing for pricing details.