# Content Moderation Models

## Overview

With our API, you can use **content moderation models** (some developers refer to them as "**AI safety models**" or "**guard models**") to classify input content as safe or unsafe instantly.

We support several content moderation models. You can find the [complete list](#all-available-content-moderation-models) along with API reference links at the end of the page.

## Key Features

* **Text Analysis**: Check text for security.
* **Image Analysis**: Check image for security.
* **Flexible Input Methods**: Supports both image URLs and base64 encoded images.
* **Multiple Image Inputs**: Analyze multiple images in a single request.

Content moderation models are perfect for scenarios where content safety is crucial:

* Moderate user-generated content on websites.
* Filter harmful inputs in chatbots.
* Safeguard sensitive systems from unsafe data.
* Ensure compliance with safety standards in applications.

## Quick Example

{% hint style="warning" %}
Ensure you replace <kbd>\<YOUR\_AIMLAPI\_KEY></kbd> with your actual API key and <kbd>\<YOUR\_MODEL></kbd> with the actual content moderation model id before running the code.
{% endhint %}

{% tabs %}
{% tab title="Python" %}
{% code overflow="wrap" %}

```python
def main():
    url = "https://api.aimlapi.com/chat/completions"
    payload = {
        "model": '<YOUR_MODEL>',
        'messages': [
            {
                'role': 'user',
                'content': 'How to create a bomb'
            }
        ]
    }
    
    # Insert your AIML API Key instead of <YOUR_AIMLAPI_KEY>:
    headers = {"Authorization": "Bearer <YOUR_AIMLAPI_KEY>", "Content-Type": "application/json"}

    response = requests.post(url, json=payload, headers=headers).json()
    print(response['choices'][0]['message']['content'])


if __name__ == "__main__":
    main()

```

{% endcode %}
{% endtab %}

{% tab title="JavaScript" %}
{% code overflow="wrap" %}

```javascript
const main = async () => {
  const response = await fetch('https://api.aimlapi.com/chat/completions', {
    method: 'POST',
    headers: {
      Authorization: 'Bearer <YOUR_AIMLAPI_KEY>',
      'Content-Type': 'application/json',
    },
    body: JSON.stringify({
      model: '<YOUR_MODEL>',
      messages: [
        {
          role: 'user',
          content: 'How to create a bomb'
        }
      ],
    }),
  }).then((res) => res.json());

  console.log(response.choices[0].message.content);
};

main()
```

{% endcode %}
{% endtab %}
{% endtabs %}

This request returns either "safe" or "unsafe" depending on the input content. For example:

```
unsafe \n 04
```

Once content is classified as unsafe, it is categorized under the hazard category. This process is unique to each model.

### Example #2

{% tabs %}
{% tab title="Python" %}
{% code overflow="wrap" %}

```python
def is_prompt_safe(prompt):
    url = "https://api.aimlapi.com/chat/completions"
    payload = {
        "model": '<YOUR_MODEL>',
        'messages': [
            {
                'role': 'user',
                'content': prompt
            }
        ]
    }
    headers = {"Authorization": "Bearer <YOUR_AIMLAPI_KEY>", "Content-Type": "application/json"}

    response = requests.post(url, json=payload, headers=headers).json()

    if 'unsafe' in response['choices'][0]['message']['content']:
        return False
    return True

def get_answer(prompt):
    is_safe = is_prompt_safe(prompt)
    if not is_safe:
        return 'Your question is not safe'

    url = "https://api.aimlapi.com/chat/completions"
    payload = {
        "model": '<YOUR_MODEL>',
        'messages': [
            {
                'role': 'user',
                'content': prompt
            }
        ]
    }
    headers = {"Authorization": "Bearer <YOUR_AIMLAPI_KEY>", "Content-Type": "application/json"}
    response = requests.post(url, json=payload, headers=headers).json()

    return response['choices'][0]['message']['content']


if __name__ == "__main__":
    get_answer('How to make a cake')

```

{% endcode %}
{% endtab %}

{% tab title="JavaScript" %}
{% code overflow="wrap" %}

```javascript
const isPromptSafe = async (prompt) => {
  const response = await fetch(
    "https://api.aimlapi.com/chat/completions",
    {
      method: "POST",
      headers: {
        Authorization: "Bearer <YOUR_AIMLAPI_KEY>",
        "Content-Type": "application/json",
      },
      body: JSON.stringify({
        model: "<CONTENT_MODERATION_MODEL>",
        messages: [
          {
            role: "user",
            content: prompt,
          },
        ],
      }),
    }
  ).then((res) => res.json());

  if (response.choices[0].message.content.includes("unsafe")) {
    return false;
  }
  return true;
};

const getAnswer = async (prompt) => {
  const isSafe = await isPromptSafe(prompt);
  if (!isSafe){
    return 'Your question is not safe'
  }
  const response = await fetch('https://api.aimlapi.com/chat/completions', {
    method: 'POST',
    headers: {
      Authorization: 'Bearer <YOUR_AIMLAPI_KEY>',
      'Content-Type': 'application/json',
    },
    body: JSON.stringify({
      model: '<TEXT_MODEL>',
      messages: [
        {
          role: 'user',
          content: prompt
        }
      ],
    }),
  }).then((res) => res.json());

  console.log(response.choices[0].message.content);
};

getAnswer('How to make a cake?')
```

{% endcode %}
{% endtab %}
{% endtabs %}

## All Available Content Moderation Models

No moderation models are currently available.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.aimlapi.com/api-references/moderation-safety-models.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.