---
title: "OCR Extraction"
method: POST
path: "/v1/ocr"
tags: ["Vision"]
---

# OCR Extraction

`POST /v1/ocr`

Extract text from images using advanced OCR models.

## Request body

- object
  - `model` string
  - `image` string, required
  - `image_type` 'url' | 'base64'
  - `return_image` boolean
  - `prompt` string
  - `async` boolean

## Response `200`

Successful OCR extraction

## Other responses

- `400` — Invalid image or model

---

[API](https://skmtc.net/neosantara/apis/neosantara-ai-api.md) · [All operations](https://skmtc.net/neosantara/apis/neosantara-ai-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/neosantara/neosantara-ai-api/revisions/82bc74ee8650/schema)