Paddle Ocr API Documentation

Paddle Ocr API Documentation

Playground

PaddleOCR-VL is an ultra-compact 0.9B parameter vision-language model for document parsing, supporting 109 languages with text, table, formula, and chart recognition in JSON or Markdown output. Ready-to-use REST inference API, best performance, no cold starts, affordable pricing.

Features

Extract text from images with WaveSpeedAI PaddleOCR — a fast, accurate optical character recognition model. Simply upload an image and get clean, structured text output in JSON or Markdown format. Perfect for document digitization, data extraction, and text recognition tasks.

Why It Works Great

• High accuracy: Powered by PaddleOCR for reliable text recognition.
• Multi-language support: Recognizes text in multiple languages.
• Flexible output: Choose between JSON or Markdown format.
• Document-friendly: Handles scanned documents, screenshots, and photos.
• Ultra-affordable: Just $0.005 per image.
• Fast processing: Quick turnaround for high-volume workflows.

Parameters

How to Use

1. Upload your image — drag and drop or paste a public URL.
2. Choose output format — select JSON for structured data or Markdown for readable text.
3. Run — click the button to process.
4. Copy or download — use the extracted text as needed.

Pricing

$0.005 per image.

Output Formats

Best Use Cases

• Document Digitization — Convert scanned documents to editable text.
• Data Extraction — Pull text from invoices, receipts, and forms.
• Screenshot Text — Extract text from screenshots and images.
• Business Cards — Digitize contact information quickly.
• Batch Processing — Process large volumes of documents affordably.
• Content Migration — Convert image-based content to text format.

Supported Content Types

• Scanned documents (PDF pages, printed text)
• Screenshots and screen captures
• Photos of documents and signs
• Handwritten text (with varying accuracy)
• Multi-column layouts
• Tables and structured content

Pro Tips for Best Results

• Use high-resolution images for better accuracy.
• Ensure good contrast between text and background.
• Straighten skewed documents before processing.
• Use JSON format when you need text positions or bounding boxes.
• Use Markdown format for clean, human-readable output.
• At $0.005 per image, batch processing is extremely cost-effective.

Notes

• If using a URL, ensure it is publicly accessible.
• Processing time is typically under a second per image.
• Accuracy depends on image quality and text clarity.
• Supports multiple languages and character sets.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "output_format": "markdown"
}
JSON
)

# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/paddle-ocr" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}"); then
  printf 'Submit failed. %s\n' "${CHECK}" >&2
  exit 1
fi

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id. %s\n' "${CHECK}" >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
  # Transient errors (429, 5xx) are retried, honoring Retry-After.
  HTTP_CODE=$(curl --silent --show-error --fail-with-body \
    --connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
    --output "${BODY_FILE}" --write-out '%{http_code}' \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
  case "${HTTP_CODE}" in
    2??) ;;
    000|429|5??)  # dropped connection or temporary error: keep polling
      printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      sleep 10; continue ;;
    *) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      cat "${BODY_FILE}" >&2; exit 1 ;;
  esac
  RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted)
      printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
      printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.

Response Parameters

Result Request Parameters

Result Response Parameters