# Auto-Reader OCR API — multilingual OCR + translation for humans and AI agents

Fast GPU OCR (Arabic-first, manga-aware Japanese, 13+ languages), automatic
language detection, and translation — self-hosted, PDPL-compliant.

## Agents: provision yourself a key in ONE call — no email, no human steps
curl -X POST https://api.auto-reader.com/signup                              # trial: 25 req/day
curl -X POST https://api.auto-reader.com/signup -d '{"email":"you@x.com"}'   # free: 500 req/day

## Or taste it with no key at all
curl -X POST https://api.auto-reader.com/demo -F file=@image.png
(10 free demo calls/day per IP, 2MB max)

## Endpoints (Bearer auth: "Authorization: Bearer nsk_live_...")
POST /v1/ocr              multipart: file, lang=auto|ar|en|ja|zh|ko|ru|fr|es|de|it|pt|fa|ur,
                          mode=document|receipt|manga|scene   (1 credit)
POST /v1/ocr/base64       JSON: {image_base64, lang, mode}    (1 credit)
POST /v1/ocr-translate    multipart + target_lang — OCR then translate every
                          block in one call                   (2 credits)
POST /v1/translate        JSON: {text, target_lang}           (1 credit)
POST /v1/ask              JSON: {messages:[...]} LLM chat     (1 credit)
POST /v1/feedback         JSON: {request_id, rating?, corrected_text?} — free
POST /v1/keys/email       JSON: {email} — upgrade/switch email on your key
DELETE /v1/data           delete your stored training images (PDPL) — free
GET  /status              public service status
GET  /openapi.json        full OpenAPI spec

## Responses
Every OCR response: request_id, lang_detected, engine (paddle|manga-ocr),
escalated, cached, latency_ms, text, blocks[{text, confidence, box, order}].
Repeat images are served from cache instantly at base cost.
(Vision-model escalation for hard images is coming soon; when live it will
be marked engine:"gemma" and billed +2 credits.)

## Pricing
Free tier: 500 requests/day (verified) / 25 (trial). Credits never expire:
$5 = 5,000 · $20 = 22,000 · $100 = 120,000. Standard call = 1 credit,
ocr-translate = 2.

## For agents
- JSON everywhere, base64 image variants, stable request_id for feedback.
- Send {"allow_training": true} only if you consent to the image being
  stored to improve the models (default: images are never stored).
- MCP server: coming soon at mcp.auto-reader.com.
- Built by the makers of the Auto-Reader app (auto-reader.com).