---
title: Image vision
---

# Image vision
POST *https://api.turing.sh/image/vision*

## Parameters


- **model**
  - Type: array
  - Required: true
  - Description: Model to use for the analysis
  - Options: blip2, ocr, gemini
  - Default: gemini

- **image**
  - Type: string
  - Required: true
  - Description: Image URL for the model to process

- **typeImage**
  - Type: string
  - Required: false
  - Description: Type of image to process
  - Options: anything, person
  - Default: anything

## Response


- **cost**
  - Type: number
  - Description: Cost of the request in USD

- **description**
  - Type: string
  - Description: Description of the image

- **text**
  - Type: string
  - Description: Text extracted from the image

- **done**
  - Type: boolean
  - Description: Whether the request is done or not

## Examples

<CodeGroup title="Image vision example">
  ```typescript
  import axios from "axios";
  (async () => {
    let response = await axios({
      method: "post",
      url: 'https://api.turing.sh/image/vision',
      headers: {
        "Content-Type": "application/json",
        Authorization: "Bearer YOUR_API_KEY",
        "x-captcha-token": "Captcha key"
      },
      data: {
  "model": [
    "gemini"
  ],
  "image": "string",
  "typeImage": "anything"
},
    })
  })();
  ```
  ```python
  import requests
  import json
  response = requests.post(
    "https://api.turing.sh/image/vision",
    headers={
      "Content-Type": "application/json",
      "Authorization": "Bearer YOUR_API_KEY",
      "x-captcha-token": "Captcha key"
    },
    data=json.dumps({
  "model": [
    "gemini"
  ],
  "image": "string",
  "typeImage": "anything"
}),
  )
  ```
</CodeGroup>