ModelInfo
English
All models

Document OCR

Text recognition for scanned or photographed documents, including tilted, low-quality and multilingual files; priced per page.

Upstageocr2026-09-30Report incorrect data

Specs

Context
—
Max output
—
Input price
$0.0015/ page
Output price
—
Released
2026-09-30

Context

Not published

Pricing · USD / page

Input
$0.0015
Output
—
Cache read
—
Cache write
—

Modalities

Input
ImageDocument
Output
Text

API

API types
ocr

Reasoning

Not published

Info

Status
Active
Released
2026-09-30
Knowledge cutoff
—

Features

Not published
Source · official docsVerified 2026-10-11

How to call

1 provider

upstagemodel = ocr

JSON

Standard format
ocr.json
{
  "id": "ocr",
  "object": "model",
  "created": 1790726400,
  "owned_by": "upstage",
  "name": "Document OCR",
  "api": {
    "types": [
      "ocr"
    ]
  },
  "limits": {
    "context": null,
    "input": null,
    "output": null
  },
  "modalities": {
    "input": [
      "image",
      "document"
    ],
    "output": [
      "text"
    ]
  },
  "reasoning": {
    "supported": null,
    "efforts": []
  },
  "pricing": {
    "currency": "USD",
    "unit": "page",
    "input": 0.0015,
    "output": null,
    "cache_read": null,
    "cache_write": null
  },
  "features": [],
  "info": {
    "status": "active",
    "release_date": "2026-09-30",
    "knowledge_cutoff": null,
    "description": "Text recognition for scanned or photographed documents, including tilted, low-quality and multilingual files; priced per page.",
    "docs": "https://console.upstage.ai/docs/models/document-ocr",
    "verified_at": "2026-10-11"
  }
}

Official sources

3