Overview

SolveOCR API extracts text from images using deep learning. Send an image, get structured text with confidence scores and bounding boxes.

Authentication

All requests require a Bearer token in the Authorization header.

Rate Limits

SolveOCR employs strict concurrency and throughput controls to ensure massive burst stability.

POST /v1/ocr

Extract text from a single image. Supports PNG, JPG, WebP, TIFF, PDF.

Example Request & Response

curl -X POST https://api.solveocr.com/v1/ocr \
  -H "Authorization: Bearer sk-..." \
  -F "image=@doc.png"

{
  "text": "Invoice #2847...",
  "confidence": 0.997,
  "language": "en"
}

POST /v1/batch

Extract text from multiple images simultaneously. Supports up to 50 concurrent images.

Example Request & Response

curl -X POST https://api.solveocr.com/v1/batch \
  -H "Authorization: Bearer sk-..." \
  -F "images=@doc1.png" \
  -F "images=@doc2.png"

{
  "results": [
    {"text": "Invoice #1...", "confidence": 0.99, "language": "en"},
    {"text": "Receipt #2...", "confidence": 0.98, "language": "en"}
  ]
}