Drake OCR API

Base URL: https://ocr-api.thedrake.ai Version: v1 Auth: API key required on protected endpoints

Authentication

Provide your API key in either of these ways:

Keep your API key secret. Rotate immediately if exposed.

Method: POST
Path: /v1/ocr/image
Auth: Required
Body: multipart/form-data
file: image file (png, jpg, jpeg)
Response: json { "text": "...markdown text..." }
Curl: bash curl -H "X-API-Key: <YOUR_API_KEY>" \ -F "file=@/path/to/image.jpg" \ https://ocr-api.thedrake.ai/v1/ocr/image

Method: POST
Path: /v1/ocr/pdf
Auth: Required
Body: multipart/form-data
file: PDF file
Response: json { "text": "...markdown text...", "pages": 3 }
Curl: bash curl -H "X-API-Key: <YOUR_API_KEY>" \ -F "file=@/path/to/file.pdf" \ https://ocr-api.thedrake.ai/v1/ocr/pdf

Method: POST
Path: /v1/export
Auth: Required
Body: application/json json { "text": "# Title\nHello", "format": "docx" } Format options: docx, pdf
Response: json { "content": "<base64>", "filename": "extracted_text.docx", "mime": "application/vnd.openxmlformats-officedocument.wordprocessingml.document" }
Save output file (bash example): bash curl -sS -H "Content-Type: application/json" \ -H "X-API-Key: <YOUR_API_KEY>" \ -d '{"text":"# Title\nHello","format":"pdf"}' \ https://ocr-api.thedrake.ai/v1/export \ | python - <<'PY' import sys, json, base64 r=json.load(sys.stdin) with open(r['filename'], 'wb') as f: f.write(base64.b64decode(r['content'])) print('Saved', r['filename']) PY