AI Image Generation

POST /v1/ai/image

Generate images from text prompts using DALL-E 3, Flux, or Stable Diffusion. Generated images can be used directly in video timelines or downloaded standalone.

Models

Model	Speed	Quality	Cost	Best For
flux	Fast (3-5s)	High	$0.14	Product shots, marketing, general purpose
stable-diffusion	Fast (2-4s)	Good	$0.14	Artistic styles, illustrations, concept art
dall-e-3	Medium (8-15s)	Highest	$0.18	Photorealism, complex scenes, text in images

Request

Field	Type	Required	Description
prompt	string	Yes	Text description of the desired image
model	string	No	dall-e-3, flux, stable-diffusion (default: flux)
width	number	No	Width in pixels (default: 1024)
height	number	No	Height in pixels (default: 1024)
style	string	No	natural, vivid, artistic (default: natural)
negative_prompt	string	No	What to exclude from the image (Flux and SD only)

Examples

curl

curl -X POST https://api.vidiking.com/v1/ai/image \
  -H "Authorization: Bearer $VIDIKING_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A modern office workspace with a laptop showing analytics dashboards, soft natural lighting, minimalist design",
    "model": "flux",
    "width": 1920,
    "height": 1080,
    "style": "natural"
  }'

Node.js

const image = await client.ai.generateImage({
  prompt: 'A modern office workspace with a laptop showing analytics dashboards, soft natural lighting',
  model: 'flux',
  width: 1920,
  height: 1080,
  style: 'natural',
});

console.log('Image URL:', image.url);

Python

image = client.ai.generate_image(
    prompt="A modern office workspace with a laptop showing analytics dashboards, soft natural lighting",
    model="flux",
    width=1920,
    height=1080,
    style="natural",
)

print("Image URL:", image.url)

Response

{
  "id": "img_abc123",
  "url": "https://cdn.vidiking.com/ai/img_abc123.png",
  "model": "flux",
  "width": 1920,
  "height": 1080,
  "credits_used": 5,
  "created_at": "2025-01-15T10:30:00Z"
}

Using Generated Images in Video

Generate an image and use it directly in a timeline.

// Generate background image
const bg = await client.ai.generateImage({
  prompt: 'Abstract gradient background in blue and purple tones',
  model: 'flux',
  width: 1920,
  height: 1080,
});

// Use in a video render
const { jobId } = await client.render({
  timeline: {
    tracks: [
      {
        clips: [
          {
            asset: { type: 'title', text: 'Welcome', style: 'minimal', size: 'large' },
            start: 0,
            length: 5,
          },
        ],
      },
      {
        clips: [
          {
            asset: { type: 'image', src: bg.url },
            start: 0,
            length: 5,
          },
        ],
      },
    ],
  },
  output: { format: 'mp4', resolution: 'hd' },
});

Model Comparison

DALL-E 3

Best for photorealistic images and complex scenes. Handles text rendering in images well. Slower and more expensive but highest quality.

curl -X POST https://api.vidiking.com/v1/ai/image \
  -H "Authorization: Bearer $VIDIKING_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A neon sign that reads OPEN 24/7 in a rainy city street at night",
    "model": "dall-e-3",
    "width": 1024,
    "height": 1024,
    "style": "vivid"
  }'

Flux

Best general-purpose model. Fast, high quality, and cost effective. Good at following detailed prompts.

curl -X POST https://api.vidiking.com/v1/ai/image \
  -H "Authorization: Bearer $VIDIKING_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Professional headshot of a confident business person, studio lighting, neutral background",
    "model": "flux",
    "width": 1024,
    "height": 1024
  }'

Stable Diffusion

Best for artistic and stylized images. Supports negative prompts for fine control.

curl -X POST https://api.vidiking.com/v1/ai/image \
  -H "Authorization: Bearer $VIDIKING_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Watercolor painting of a coastal village at sunset, vibrant colors",
    "model": "stable-diffusion",
    "width": 1024,
    "height": 1024,
    "style": "artistic",
    "negative_prompt": "blurry, low quality, distorted"
  }'

Supported Sizes

Aspect Ratio	Dimensions
1:1	512x512, 1024x1024
16:9	1024x576, 1920x1080
9:16	576x1024, 1080x1920
4:3	1024x768
3:4	768x1024