* * * OCR FOR AGENTS * * *

OCR Built for AI Agents.

Agents don't read dashboards — they need tools that state their price, enforce their caps, and fail with errors they can recover from. PennyOCR ships a remote MCP server (mcp.pennyocr.com), a free cost estimator, hard max_pages/max_cost_usd guards, prepaid credits that bound worst-case spend, and a signup flow documented for non-humans at /llms.txt.

START FREE — 50 PAGES
$0.75 / 1,000 PAGES
1,333 PAGES PER DOLLAR · NO CREDIT CARD
AT A GLANCE
UPDATED
PRICE
$0.75 per 1,000 pages ($0.00075 per page), prepaid
FREE TIER
First 50 pages free per account, no credit card
FORMATS
PDF, PNG, JPEG, WebP, TIFF
LIMITS
50 MB and 500 pages per request
OUTPUT
JSON with markdown (tables preserved) or plain text, per page and joined
INTERFACES
REST API, OpenAI-compatible chat completions, MCP server
CREDITS
Never expire; requests stop with a 402 when the balance runs out
01 / TRY IT

ONE ENDPOINT.

POST a file, get JSON back — the extracted text, per page and joined. PDF, PNG, JPEG, WebP or TIFF.

curl https://api.pennyocr.com/v1/ocr \
  -H "Authorization: Bearer $PENNYOCR_API_KEY" \
  -F "file=@report.pdf"

# $0.75 per 1,000 pages, first 50 free
02 / USE CASES

AGENT-NATIVE BY DESIGN.

MCP SERVER
Add {"url": "https://mcp.pennyocr.com"} to mcpServers and your agent sees read_document and estimate_cost. Works with Claude, OpenAI Agents SDK, Cloudflare Agents.
COST BEFORE COMMIT
estimate_cost is free and exact: '200 pages, $0.15'. The agent can reason about spend before spending.
HARD CAPS
max_pages and max_cost_usd refuse deterministically — with the numbers — before any GPU work. A looping agent stops at your prepaid balance, never your card limit.
RECOVERABLE ERRORS
401/402/413/422 with machine-readable detail strings that say exactly what to change. No mystery 500s to retry into.
03 / PRICE CHECK

HALF THE PRICE OF THE BIG CLOUDS.

Per 1,000 pages, public list prices, first tier.

PENNYOCR$0.75
AWS TEXTRACT$1.50
GOOGLE CLOUD VISION$1.50
AZURE DOC INTELLIGENCE$1.50
YOU KEEP50%
04 / FAIR QUESTIONS
GET YOUR API KEY →
QUESTIONS? HELLO@PENNYOCR.COM
HOW DOES AN AGENT GET A KEY?
The whole flow is REST and documented at pennyocr.com/llms.txt: request an email code, verify it, mint a key, optionally buy credits via a Stripe checkout URL a human completes.
IS THERE AN OPENAI-COMPATIBLE ENDPOINT?
Yes — POST /v1/chat/completions with image_url parts, model pennyocr-markdown or pennyocr-text. Any OpenAI SDK works by changing base_url.
WHAT ABOUT GIANT PDFS?
Page ranges today (pages: '1-20'), up to 500 pages per request. Async jobs with webhooks are on the roadmap.