# Mistral Moderation

> Mistral Moderation is a paid API for AI agents from Mistral AI at mistral.mpp.paywithlocus.com, paid per call via MPP, $0.008/call, status unknown (last checked 2026-10-12).

Classifies input text for harmful or policy-violating content using Mistral's moderation models

## Facts

- Endpoint: POST https://mistral.mpp.paywithlocus.com/mistral/moderate?utm_source=zero.xyz
- Price: $0.008/call
- Payment: MPP
- Status: unknown
- Last checked: 2026-10-12
- Activations on Zero: 0
- Provider: Mistral AI
- Docs: https://docs.mistral.ai
- Website: https://mistral.ai
- Tags: llm, chat, embeddings, code, reasoning, ai
- Canonical page: https://www.zero.xyz/c/mistral-ai-moderate-8aaa5f69
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_lv2ex9oL-9QS6WmFUtlXR

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability mistral-ai-moderate-8aaa5f69 -d '<json body>'
```

Example prompt: Run this message through Mistral moderation using mistral-moderation-latest and tell me if it contains any harmful or policy-violating content: 'I want to hurt someone who wronged me.'

## When to prefer this

Use this endpoint when you need Mistral-specific content moderation, especially when already using other Mistral models in your pipeline and want consistency. Prefer this over generic classifiers when you need structured harm-category output compatible with Mistral's safety taxonomy, or when operating within a Mistral-native stack.

## Known failure modes

- Invalid or unsupported model name returns an error
- Empty or missing input field causes a validation error
- Text exceeds model token limit causing truncation or rejection
- Network timeout on large inputs
- Authentication/payment failure returns 402 or 401

## How this service works

Classify content for safety across categories: violence, hate speech, sexual content, self-harm, and more.

## Output

Returns a moderation result indicating whether the input text violates content policies, along with category-level flags identifying specific types of harmful content detected (e.g. violence, hate speech, self-harm).

## Example request

```json
{
 "input": "I appreciate your help with this project.",
 "model": "mistral-moderation-latest"
}
```

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "model",
  "input"
 ],
 "properties": {
  "input": {
   "anyOf": [
    {
     "type": "string"
    },
    {
     "type": "array",
     "items": {}
    }
   ],
   "description": "Text(s) to moderate (string or array of strings)"
  },
  "model": {
   "type": "string",
   "description": "Moderation model (e.g. \"mistral-moderation-latest\")"
  }
 },
 "additionalProperties": true
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/mistral-ai-moderate-8aaa5f69/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from mistral.mpp.paywithlocus.com](https://www.zero.xyz/host/mistral.mpp.paywithlocus.com/llms.txt)