API Endpoints
Image Generation API
Image Generation API
Generate high-quality images and professional logos using various AI models.
Image Generation Endpoint
Code
Request Body
| Field | Type | Required | Description |
|---|---|---|---|
prompt | string | Yes | Text description of the desired image (1-4000 chars) |
model | string | Yes | The AI model to use |
size | string | No | Image dimensions (default: "square") |
n | integer | No | Number of images to generate (1-4, default: 1) |
negative_prompt | string | No | What to avoid in the image |
style | string | No | Artistic style preset |
quality | string | No | Image quality ("standard" or "hd") |
seed | integer | No | Seed for reproducible generation |
guidance_scale | number | No | Prompt adherence (1-20) |
num_inference_steps | integer | No | Denoising steps (10-150) |
Available Models
Premium Models
dall-e-3- OpenAI's DALL-E 3gpt-image-1- GPT Image modelfal-ai/flux-pro/v1.1-ultra- Flux Pro Ultrafal-ai/flux-pro/v1.1- Flux Pro
Standard Models
fal-ai/flux/schnell- Fast Flux modelfal-ai/flux/dev- Flux Developmentfal-ai/stable-diffusion-v3-medium- SD3 Mediumfal-ai/stable-diffusion-v35-large/turbo- SD3.5 Turbofal-ai/stable-diffusion-v35-large- SD3.5 Largefal-ai/sana/sprint- Sana Sprint
Size Options
square- 1024x1024square_hd- 1024x1024 HDsquare_small- 512x512landscape_4_3- 1024x768landscape_16_9- 1024x576portrait_4_3- 768x1024portrait_16_9- 576x1024
Style Options
photorealistic,anime,digital_artoil_painting,watercolor,sketchcartoon,abstract,3d_renderpixel_art,minimalist,surrealvintage,retro,neonfantasy,sci-fi,steampunkcyberpunk,gothic
Example Request
Code(bash)
Response Format
Code(json)
Logo Generation Endpoint
Code
Request Body
| Field | Type | Required | Description |
|---|---|---|---|
prompt | string | Yes | Logo description (1-2000 chars) |
model | string | Yes | The AI model to use |
style | string | Yes | Logo style |
industry | string | Yes | Industry category |
size | string | No | Logo dimensions |
n | integer | No | Number of variations (1-4) |
color_scheme | string | No | Preferred colors |
typography | string | No | Font style preference |
mood | string | No | Logo mood/feeling |
complexity | integer | No | Design complexity (1-5) |
refine_prompt | boolean | No | AI prompt refinement |
Logo Styles
minimal- Clean and simplemodern- Contemporary designvintage- Classic/retro lookplayful- Fun and casualcorporate- Professional businesstech- Technology-focusedorganic- Natural/flowinggeometric- Shape-basedhand-drawn- Artistic/sketchedabstract- Non-representational
Industries
technology,healthcare,financeeducation,retail,foodtravel,entertainment,sportsreal-estate,automotive,fashionbeauty,fitness,consultinglegal,construction,agricultureenergy,nonprofit
Typography Options
sans-serif- Clean, modernserif- Traditional, elegantscript- Flowing, decorativedisplay- Bold, attention-grabbingmonospace- Technical, uniform
Mood Options
professional- Serious, business-likefriendly- Approachable, warmbold- Strong, confidentelegant- Sophisticated, refinedplayful- Fun, energetic
Logo Example
Code(json)
Credit Usage
Credit consumption varies by model and options:
Image Generation
- 1 credit per image
Logo Generation
- 1 credit per logo
Multiple images (n > 1) multiply the credit cost.
Advanced Features
Negative Prompts
Use negative prompts to exclude unwanted elements:
Code(json)
Reproducible Generation
Use seeds for consistent results:
Code(json)
Fine-tuning Parameters
For Stable Diffusion and Flux models:
Code(json)
Best Practices
-
Prompt Engineering
- Be specific and descriptive
- Include style, mood, and lighting
- Mention important details first
-
Model Selection
- Flux Schnell: Fast, good quality
- DALL-E 3: Best for creative/artistic
- Stable Diffusion: Great control options
-
Optimization
- Start with lower quality for testing
- Use appropriate sizes
- Generate single images first
-
Logo Design
- Always specify industry and style
- Include brand personality
- Consider multiple variations
Error Handling
Model Not Available
Code(json)
Invalid Parameters
Code(json)
Image Storage
Generated images are:
- Stored temporarily (30 days)
- Accessible via secure URLs
- Can be downloaded immediately
- Not associated with your account after generation
Rate Limits
Image generation has specific rate limits:
- Requests per minute vary by plan
- Concurrent generation limits apply
- Premium models may have lower limits
Next Steps
Last modified on