Image Ninja API

Remove background from images using AI.

Parameters

image	File (required)	Image file (PNG, JPG, WebP)
output	String	"json" (default) or "raw" for binary image

Example

curl -X POST "https://image-ninja.io/api/v1/remove-background" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]"

Response

{
  "success": true,
  "data": {
    "image": "base64_encoded_image...",
    "mime_type": "image/png",
    "width": 800,
    "height": 600,
    "processing_time_ms": 1234
  }
}

Generate images from text descriptions using AI (Stable Diffusion XL via ClipDrop).

Parameters

prompt	String (required)	Text description of the image to generate (max 1000 chars)
output	String	"json" (default) or "raw" for binary image

Example

curl -X POST "https://image-ninja.io/api/v1/text-to-image" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "prompt=A serene mountain landscape at sunset with a calm lake"

Response

{
  "success": true,
  "data": {
    "image": "base64_encoded_image...",
    "mime_type": "image/png",
    "width": 1024,
    "height": 1024,
    "prompt": "A serene mountain landscape..."
  }
}

Extend images beyond their original boundaries using AI. Perfect for changing aspect ratios or adding more context to cropped images.

Parameters

image	File (required)	Image file (PNG, JPG, WebP)
extend_left	Integer	Pixels to extend on the left (0-1024)
extend_right	Integer	Pixels to extend on the right (0-1024)
extend_up	Integer	Pixels to extend upward (0-1024)
extend_down	Integer	Pixels to extend downward (0-1024)

Example

curl -X POST "https://image-ninja.io/api/v1/image-uncrop" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" \
  -F "extend_left=200" \
  -F "extend_right=200"

Response

{
  "success": true,
  "data": {
    "image": "base64_encoded_image...",
    "mime_type": "image/png",
    "original_width": 800,
    "original_height": 600,
    "width": 1200,
    "height": 600,
    "extend_left": 200,
    "extend_right": 200
  }
}

Upscale images 2x or 4x using AI while preserving detail and quality.

Parameters

image	File (required)	Image file (PNG, JPG, WebP)
scale	Integer	2 or 4 (default: 2)

Example

curl -X POST "https://image-ninja.io/api/v1/upscale" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" \
  -F "scale=2"

Response

{
  "success": true,
  "data": {
    "image": "base64_encoded_image...",
    "mime_type": "image/jpeg",
    "original_width": 400,
    "original_height": 300,
    "width": 800,
    "height": 600,
    "scale": 2
  }
}

Remove unwanted objects from images using AI inpainting. Uses ClipDrop with LaMa fallback for best results.

Parameters

image	File (required)	Image file (PNG, JPG, WebP)
mask	String (required)	Base64-encoded mask image (white = areas to remove)
engine	String	"auto" (default), "clipdrop", or "lama"

Example

curl -X POST "https://image-ninja.io/api/v1/magic-eraser" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" \
  -F "mask=data:image/png;base64,..."

Response

{
  "success": true,
  "data": {
    "image": "base64_encoded_image...",
    "mime_type": "image/png",
    "width": 800,
    "height": 600,
    "engine": "clipdrop"
  }
}

Automatically remove all text, watermarks, captions, and overlays from images using AI.

Parameters

image

File (required)

Image file (PNG, JPG, WebP)

Example

curl -X POST "https://image-ninja.io/api/v1/remove-text" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "image=@photo_with_text.jpg"

Response

{
  "success": true,
  "data": {
    "image": "base64_encoded_image...",
    "mime_type": "image/png",
    "width": 800,
    "height": 600
  }
}

Compress images while maintaining quality.

Parameters

image	File (required)	Image file
quality	Integer	1-100 (default: 80)
format	String	jpg, png, webp

Response

{
  "success": true,
  "data": {
    "image": "base64...",
    "original_size": 245000,
    "compressed_size": 89000,
    "savings_percent": 63.7
  }
}

Generate QR codes from text or URLs.

Parameters

content	String (required)	Text or URL to encode
size	Integer	50-2000 (default: 300)
foreground	String	Hex color (default: #000000)
background	String	Hex color (default: #FFFFFF)

Get company logo for any domain.

Example

curl "https://image-ninja.io/api/v1/logo/google.com" \
  -H "Authorization: Bearer YOUR_API_KEY"

Response

{
  "success": true,
  "data": {
    "domain": "google.com",
    "logo": "base64...",
    "favicon_url": "https://...",
    "brand_colors": ["#4285F4", "#34A853", "#FBBC05", "#EA4335"]
  }
}

Extract text from images using OCR.

Parameters

image	File (required)	Image file
language	String	eng, spa, fra, deu, ita, etc.

Response

{
  "success": true,
  "data": {
    "text": "Extracted text content...",
    "word_count": 42,
    "line_count": 5,
    "language": "eng"
  }
}

Parameters

image	File (required)	Image file (PNG, JPG, WebP)
task	String (required)	caption, ocr, object_detection, dense_caption, grounding, open_detection
detail_level	String	For caption: short, detailed, very_detailed
with_regions	Boolean	For ocr: include text coordinates
query	String	For grounding: what to find (e.g., "the red car")
objects	String	For open_detection: comma-separated list (e.g., "cat, dog, car")

Credit Costs

caption	1 credit	Generate image descriptions
ocr	1 credit	Extract text from images
object_detection	2 credits	Detect all objects with bounding boxes
dense_caption	2 credits	Describe regions of the image
grounding	2 credits	Find objects by text description
open_detection	2 credits	Detect specific named objects

Example: Image Captioning

curl -X POST https://image-ninja.io/api/v1/florence \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" \
  -F "task=caption" \
  -F "detail_level=detailed"

Response (Caption)

{
  "success": true,
  "data": {
    "task": "caption",
    "caption": "A golden retriever sitting on a green lawn in a sunny park, with trees in the background.",
    "detail_level": "detailed",
    "processing_time_ms": 1234,
    "image_size": {"width": 1920, "height": 1080}
  }
}

Example: Object Detection

curl -X POST https://image-ninja.io/api/v1/florence \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" \
  -F "task=object_detection"

Response (Object Detection)

{
  "success": true,
  "data": {
    "task": "object_detection",
    "objects": [
      {"label": "car", "bbox": {"x1": 100, "y1": 200, "x2": 400, "y2": 500}},
      {"label": "person", "bbox": {"x1": 450, "y1": 150, "x2": 550, "y2": 600}}
    ],
    "total_objects": 5,
    "object_counts": {"car": 2, "person": 3},
    "unique_labels": ["car", "person"],
    "processing_time_ms": 1567
  }
}

Example: Visual Grounding

curl -X POST https://image-ninja.io/api/v1/florence \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" \
  -F "task=grounding" \
  -F "query=the red sofa"

Response (Grounding)

{
  "success": true,
  "data": {
    "task": "grounding",
    "query": "the red sofa",
    "found": true,
    "match_count": 1,
    "matches": [
      {"phrase": "the red sofa", "bbox": {"x1": 50, "y1": 300, "x2": 600, "y2": 550}}
    ],
    "processing_time_ms": 1123
  }
}

Get current account information and credit balance.

Get API usage statistics.

Parameters

range

String

7d, 30d, 90d, all (default: 30d)