Convert images to text, tables, and structured data with a single image OCR API. Pay-as-you-go at $1 per 500 calls — no subscription, your balance never expires. Sign up and get $0.10 in free API credit for debugging. $1 = 500 API calls API calls per dollar Requests per minute API keys per account Three APIs in one platform One API key unlocks text extraction, table digitization, and structured field extraction — each returns ready-to-use JSON, CSV, XLSX, or DOCX. Image to Text API Extract every word from an image with layout preservation. Input: Images (PNG, JPEG, WEBP) Output: Markdown, plain text, or DOCX Detect and digitize tables from images into structured rows. Input: Images with tables (PNG, JPEG, WEBP) Output: JSON (title + HTML), CSV, or XLSX Data Extraction API Pull specific fields from invoices, receipts, and forms. Input: Images + field names Output: JSON (key + values), CSV, or XLSX OCR REST API Reference & Live Examples Every example below is copy-ready and matches the current API. Authenticate once, then call any endpoint. Extract text from images with layout preservation. Output as Markdown (default), plain text, or DOCX. Detect and extract tables from images into structured JSON, CSV, or XLSX formats. Extract specific fields from images using AI. Perfect for invoices, receipts, and forms. Authenticate with one header Pass your API key as a Bearer token in the Authorization header on every request. Up to 5 API keys per account. Supported formats & limits Send images as multipart/form-data with an image field. Each endpoint accepts the same input. Image files — PNG, JPEG, WEBP (multipart image field) image (required) 200 requests per minute per key by default (adjustable on request). Images are auto-resized to a maximum of 2048×2048 px before processing. PDF is not supported by the REST API; use the web or desktop app for PDFs. Error handling & billing Only successful calls are billed. You never pay for a request that errors out. $1 = 500 calls, deducted from your balance per successful call. If processing fails, the charge is automatically refunded to your balance. Authentication, validation, rate-limit, and insufficient-balance errors incur no charge. Your balance is shown in dollars and updates in real time. Unauthorized — invalid or missing API key Bad request — no image file (or no fields for /extract) Insufficient balance — please top up

Rate limit exceeded — try again later Internal error — processing failed (charge refunded) Pay-as-you-go, no subscription No hidden fees. No subscriptions. Just pay for what you use. 500 API calls per $1 $5 minimum top-up Balance never expires Balance is measured in dollars — top up from $5 and watch it go down only on success. Built for real workloads From document processing pipelines to mobile app integration, the OCR API fits anywhere images need to become data. Document processing Turn scanned documents and screenshots into searchable Markdown or DOCX at scale. Invoice & receipt extraction Pull invoice numbers, dates, totals, and line items into structured fields automatically. Table digitization Convert photographed spreadsheets and printed tables into CSV or XLSX in one call. App integration Embed image-to-text and data extraction into your web, mobile, or desktop product. Previously using RapidAPI? Your privacy matters Privacy-first processing API content is used only to provide OCR and data extraction. We minimize retention and do not use your content for advertising or unrelated purposes. Necessary operational data We process only the account and technical data needed for billing, security, and reliable API operation. Secure transfer All API requests are transmitted over secure HTTPS connections to ensure your data is protected during transit. Frequently asked questions Is the OCR API free, and how much does it cost? New accounts receive $0.10 in free API credit for 50 successful calls. After that, the OCR API costs $1 per 500 successful calls. There is no subscription, the balance never expires, and the minimum top-up is $5. What file formats are supported? The API accepts image files (PNG, JPEG, WEBP) sent as a multipart image field. Images are auto-resized to a maximum of 2048×2048 px before processing. PDF is not supported by the REST API — use the web or desktop app for PDFs. What is the difference between OCR, Table and Extraction APIs? The OCR API returns all text from an image as Markdown, plain text, or DOCX. The Table Extraction API detects tables and returns them as JSON (title + HTML), CSV, or XLSX. The Data Extraction API takes field names you specify and returns the matching values as JSON, CSV, or XLSX — ideal for invoices and forms. Are failed requests charged? No. You're billed only for successful calls. If processing fails, the charge is automatically refunded to your balance. Authentication, validation, rate-limit, and insufficient-balance errors incur no charge at all. Does the API balance expire? No. Your API balance never expires. It is shown in dollars and remains available until you use it. Can the API process PDFs? No. The REST API processes image files only (PNG, JPEG, WEBP). For PDF to text or PDF to Word, use the FreeOCR.AI web or desktop app, which handles PDF pages directly. Is there a request size or rate limit? The default rate limit is 200 requests per minute per API key, and can be raised for high-volume users on request. The total request body must not exceed 4.5 MB due to Vercel's request size limit. Images are automatically resized to a maximum of 2048×2048 px before processing. Ready to build with the OCR API? Create your account, grab an API key, and start debugging with 50 free successful calls. Top up from $5 when you're ready.