Fusion AI Logo

Fusion AI

Documentation

← Back to Fusion AI

Image & PDF Input

Powerful multimedia processing capabilities that let AI models analyze images, extract text from PDFs, and understand visual content with ease.

Images

Supported Formats

JPEGPNGWebPGIF

Max Size: 20MB

Visual analysis and description
OCR text extraction
Object and scene recognition
Chart and diagram interpretation

Documents

Supported Formats

PDFTextMarkdown

Max Size: 50MB

Full text extraction
Document structure analysis
Table and form recognition
Multi-page processing

How to Upload Files

Base64 Encoding

{
  "prompt": "Analyze this image",
  "provider": "neuroswitch",
  "files": [
    {
      "type": "image",
      "data": "data:image/jpeg;base64,/9j/4AAQ...",
      "filename": "chart.jpg"
    }
  ]
}

URL Reference

{
  "prompt": "Extract text from this PDF",
  "provider": "neuroswitch",
  "files": [
    {
      "type": "pdf",
      "url": "https://example.com/document.pdf",
      "filename": "contract.pdf"
    }
  ]
}

Common Use Cases

Document Analysis

Extract insights from PDFs, forms, and documents

Invoice processing and data extraction
Contract review and summarization
Research paper analysis
Legal document review

Visual Understanding

Analyze images, charts, and visual content

Chart and graph interpretation
Medical image analysis
Product photo descriptions
Scene and object recognition

OCR & Text Extraction

Convert images and PDFs to searchable text

Scanned document digitization
Handwritten text recognition
Receipt and invoice parsing
Sign and label reading

Processing Examples

Image Analysis

curl -X POST https://api.mcp4.ai/chat \
  -H "Authorization: Bearer sk-fusion-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Describe what you see in this image and identify any text",
    "provider": "neuroswitch",
    "files": [
      {
        "type": "image", 
        "data": "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD...",
        "filename": "screenshot.jpg"
      }
    ]
  }'

PDF Text Extraction

{
  "prompt": "Extract all the key information from this invoice",
  "provider": "neuroswitch", 
  "files": [
    {
      "type": "pdf",
      "url": "https://storage.example.com/invoice-2024-001.pdf",
      "filename": "invoice.pdf"
    }
  ],
  "max_tokens": 1000
}

Response Format

{
  "response": "This image shows a bar chart displaying quarterly sales data...",
  "provider_used": "claude-3-opus",
  "file_analysis": {
    "files_processed": 1,
    "total_pages": 3,
    "extracted_text_length": 2847,
    "processing_time_ms": 1250
  },
  "tokens_used": 456,
  "cost": 0.00234
}

Provider Support Matrix

FeatureGPT-4VClaude 3Gemini Pro
Image Analysis✅ Excellent✅ Excellent✅ Good
PDF Processing✅ Native✅ Native🔄 Converted
OCR Accuracy⭐⭐⭐⭐⭐⭐⭐⭐⭐⭐⭐⭐⭐⭐
Max File Size20MB20MB20MB

Security & Privacy

Files encrypted in transit and at rest
Automatic file deletion after processing
No permanent storage of uploaded content
GDPR and SOC 2 compliant processing

Best Practices

✅ Optimization Tips

Use high-resolution images for better OCR accuracy
Compress large files to reduce processing time
Provide specific prompts for better analysis
Use NeuroSwitch for optimal model selection

⚠️ Considerations

Processing time increases with file size
Complex documents may require multiple requests
Handwritten text recognition has lower accuracy
Some providers have different strengths

Integration Examples

Quick Start

Try multimedia processing with your first API call

Tool Calling

Combine file processing with function calls

Advanced Config

Fine-tune processing parameters

Start Processing Multimedia

Upload images and PDFs to unlock powerful AI analysis capabilities. Get started with your first multimedia request today.

Try It NowView API Reference