Skip to main content
curl -X POST "https://api.tensorone.ai/v2/ai/image-upscale" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "imageUrl": "https://example.com/low-res-photo.jpg",
    "upscaleFactor": 4,
    "model": "real-esrgan",
    "enhanceMode": "high_quality",
    "denoiseStrength": 0.6,
    "outputFormat": "png"
  }'

{
  "upscaleId": "ups_abc123_def456",
  "status": "processing",
  "originalImage": {
    "width": 512,
    "height": 384,
    "format": "jpg",
    "fileSize": "245KB"
  },
  "processingDetails": {
    "model": "real-esrgan",
    "upscaleFactor": 4,
    "estimatedTime": "45-60 seconds"
  },
  "createdAt": "2024-01-16T15:30:00Z"
}
Enhance image resolution and quality using advanced AI upscaling models. Perfect for improving low-resolution images, old photos, or generated content that needs higher quality.

​
Request Body

​
imageUrl
string
Public URL of the image to upscale (HTTPS required)
​
imageBase64
string
Base64-encoded image data (alternative to imageUrl)
​
upscaleFactor
integer
default:"2"
Upscaling factor: 2 (2x), 4 (4x), or 8 (8x) the original resolution
​
model
string
default:"real-esrgan"
Upscaling model to use:
  • real-esrgan - Best for general images and photos
  • real-esrgan-anime - Optimized for anime/cartoon images
  • esrgan-plus - Enhanced model for photorealistic images
  • waifu2x - Specialized for anime artwork and illustrations
  • srcnn - Fast upscaling with good detail preservation
​
enhanceMode
string
default:"standard"
Enhancement mode:
  • standard - Balanced quality and speed
  • high_quality - Maximum quality, slower processing
  • fast - Quick processing, good for previews
  • face_enhance - Specialized for faces and portraits
​
denoiseStrength
number
default:"0.5"
Noise reduction strength (0.0 to 1.0). Higher values remove more noise but may reduce detail.
​
preserveDetails
boolean
default:"true"
Whether to preserve fine details and textures during upscaling
​
outputFormat
string
default:"png"
Output image format: png, jpg, webp
​
jpegQuality
integer
default:"95"
JPEG quality (1-100, only used when outputFormat is ‘jpg’)

​
Response

​
upscaleId
string
Unique identifier for the upscaling job
​
status
string
Processing status: queued, processing, completed, failed
​
originalImage
object
Information about the original image
​
upscaledImage
object
Information about the upscaled image (available when completed)
​
processingDetails
object
Processing information

​
Example

curl -X POST "https://api.tensorone.ai/v2/ai/image-upscale" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "imageUrl": "https://example.com/low-res-photo.jpg",
    "upscaleFactor": 4,
    "model": "real-esrgan",
    "enhanceMode": "high_quality",
    "denoiseStrength": 0.6,
    "outputFormat": "png"
  }'

{
  "upscaleId": "ups_abc123_def456",
  "status": "processing",
  "originalImage": {
    "width": 512,
    "height": 384,
    "format": "jpg",
    "fileSize": "245KB"
  },
  "processingDetails": {
    "model": "real-esrgan",
    "upscaleFactor": 4,
    "estimatedTime": "45-60 seconds"
  },
  "createdAt": "2024-01-16T15:30:00Z"
}

{
  "upscaleId": "ups_abc123_def456",
  "status": "completed",
  "originalImage": {
    "width": 512,
    "height": 384,
    "format": "jpg",
    "fileSize": "245KB"
  },
  "upscaledImage": {
    "url": "https://upscaled.tensorone.ai/ups_abc123_def456.png",
    "width": 2048,
    "height": 1536,
    "format": "png",
    "fileSize": "8.2MB",
    "downloadUrl": "https://upscaled.tensorone.ai/download/ups_abc123_def456?expires=..."
  },
  "processingDetails": {
    "model": "real-esrgan",
    "upscaleFactor": 4,
    "processingTime": 52.3
  },
  "createdAt": "2024-01-16T15:30:00Z",
  "completedAt": "2024-01-16T15:30:52Z"
}

​
Specialized Use Cases

​
Photo Restoration

Restore old or damaged photos: 
restoration_response = requests.post(
    "https://api.tensorone.ai/v2/ai/image-upscale",
    json={
        "imageUrl": "https://example.com/old-photo.jpg",
        "upscaleFactor": 2,
        "model": "real-esrgan",
        "enhanceMode": "face_enhance",  # Good for portraits
        "denoiseStrength": 0.8,  # Remove artifacts from old photos
        "preserveDetails": True
    }
)

​
Anime/Artwork Enhancement

Optimize for anime and illustrations: 
anime_response = requests.post(
    "https://api.tensorone.ai/v2/ai/image-upscale",
    json={
        "imageUrl": "https://example.com/anime-art.png",
        "upscaleFactor": 4,
        "model": "real-esrgan-anime",  # Specialized for anime
        "enhanceMode": "high_quality",
        "denoiseStrength": 0.3,  # Lower noise reduction for clean art
        "outputFormat": "png"  # Preserve transparency
    }
)

​
Batch Processing

Upscale multiple images: 
image_urls = [
    "https://example.com/image1.jpg",
    "https://example.com/image2.jpg", 
    "https://example.com/image3.jpg"
]

upscale_jobs = []

for url in image_urls:
    response = requests.post(
        "https://api.tensorone.ai/v2/ai/image-upscale",
        headers={"Authorization": "Bearer YOUR_API_KEY"},
        json={
            "imageUrl": url,
            "upscaleFactor": 2,
            "model": "real-esrgan",
            "enhanceMode": "fast"  # Faster for batch processing
        }
    )
    
    upscale_jobs.append(response.json()['upscaleId'])

print(f"Started {len(upscale_jobs)} upscaling jobs")

# Monitor all jobs
def check_all_jobs(job_ids):
    completed = []
    for job_id in job_ids:
        status_response = requests.get(
            f"https://api.tensorone.ai/v2/ai/upscale/{job_id}/status",
            headers={"Authorization": "Bearer YOUR_API_KEY"}
        )
        status = status_response.json()
        
        if status['status'] == 'completed':
            completed.append(status)
    
    return completed

# Wait for all to complete
import time
while len(check_all_jobs(upscale_jobs)) < len(upscale_jobs):
    print(f"Completed: {len(check_all_jobs(upscale_jobs))}/{len(upscale_jobs)}")
    time.sleep(15)

print("All upscaling jobs completed!")

​
Quality Comparison

Compare results from different models: 
def compare_models(image_url):
    models = ['real-esrgan', 'real-esrgan-anime', 'esrgan-plus', 'waifu2x']
    results = {}
    
    for model in models:
        response = requests.post(
            "https://api.tensorone.ai/v2/ai/image-upscale",
            headers={"Authorization": "Bearer YOUR_API_KEY"},
            json={
                "imageUrl": image_url,
                "upscaleFactor": 2,
                "model": model,
                "enhanceMode": "standard"
            }
        )
        
        results[model] = response.json()['upscaleId']
    
    return results

# Compare different models on same image
comparison = compare_models("https://example.com/test-image.jpg")
print("Model comparison started:", comparison)

​
Advanced Options

​
Custom Enhancement Pipeline

Chain multiple enhancements: 
# Step 1: Upscale
upscale_response = requests.post(
    "https://api.tensorone.ai/v2/ai/image-upscale",
    json={
        "imageUrl": "https://example.com/image.jpg",
        "upscaleFactor": 2,
        "model": "real-esrgan"
    }
)

upscale_id = upscale_response.json()['upscaleId']

# Wait for completion, then enhance further
# ... wait for completion ...

# Step 2: Apply additional enhancement
enhance_response = requests.post(
    "https://api.tensorone.ai/v2/ai/image-enhance",
    json={
        "imageUrl": completed_upscale['upscaledImage']['url'],
        "enhanceType": "detail_boost",
        "strength": 0.7
    }
)

​
Face-Specific Enhancement

Enhanced processing for portraits: 
face_enhance_response = requests.post(
    "https://api.tensorone.ai/v2/ai/image-upscale",
    json={
        "imageUrl": "https://example.com/portrait.jpg",
        "upscaleFactor": 4,
        "model": "real-esrgan",
        "enhanceMode": "face_enhance",
        "faceEnhancement": {
            "enabled": True,
            "skinSmoothing": 0.3,
            "eyeSharpening": 0.8,
            "detailPreservation": 0.9
        }
    }
)

​
Supported Image Formats

​
Input Formats

  • JPEG/JPG: Universal support, good for photos
  • PNG: Transparency preserved, good for graphics
  • WebP: Modern format with good compression
  • BMP: Basic bitmap format
  • TIFF: High-quality format support

​
Output Formats

  • PNG: Best quality, supports transparency
  • JPEG: Smaller file size, good for photos
  • WebP: Modern format with excellent compression

​
Limitations and Guidelines

​
File Size Limits

  • Maximum input size: 50MB per image
  • Maximum resolution: 16K x 16K pixels for input
  • Processing time: Varies by size and upscale factor

​
Best Practices

  • Choose the right model: Real-ESRGAN for photos, Waifu2x for anime
  • Start with 2x upscaling: Test before going to 4x or 8x
  • Consider file size: Higher upscale factors create much larger files
  • Format selection: Use PNG for quality, JPEG for smaller files
  • Noise reduction: Adjust based on input image quality

​
Pricing

  • 2x Upscaling: $0.05 per megapixel of output
  • 4x Upscaling: $0.10 per megapixel of output
  • 8x Upscaling: $0.20 per megapixel of output
  • Face Enhancement: Additional $0.02 per face detected
  • Batch Processing: 10% discount for 10+ images
Processing time depends on image size and upscale factor. Typical times: 2x (10-30s), 4x (30-90s), 8x (60-180s).
Very large upscale factors (8x) can create files over 100MB. Consider your storage and bandwidth requirements when choosing upscale factors.