# MistralAI-Training

> Builds datasets for training Mistral's generative models. Not used for search indexing or live answers.

Canonical page: https://getrankroot.com/ai-bots/mistralai-training · Facts checked against Mistral AI docs on 2026-10-20

| Operator | Used for | Type | Obeys robots.txt |
| --- | --- | --- | --- |
| Mistral AI | Mistral model training | AI training crawler | Yes |

## Should you allow it?

Your choice.

**If you block it:** Your content is not collected for Mistral's training datasets.

## robots.txt rules

Allow MistralAI-Training:

```
User-agent: MistralAI-Training
Allow: /
```

Block MistralAI-Training:

```
User-agent: MistralAI-Training
Disallow: /
```

Disallow MistralAI-Training from one folder only:

```
User-agent: MistralAI-Training
Disallow: /private/
Allow: /
```

A bot follows the most specific User-agent group that names it and ignores the generic `User-agent: *` group, so repeat any shared Disallow lines inside it.

## Same company, different bots: Mistral AI

| Bot | Type | Purpose | Obeys robots.txt | If you block it |
| --- | --- | --- | --- | --- |
| [MistralAI-User](https://getrankroot.com/ai-bots/mistralai-user.md) | User-triggered fetcher | Mistral Vibe (user requests) | Yes | Mistral's assistant cannot visit your pages to answer users' questions. |
| [MistralAI-Index](https://getrankroot.com/ai-bots/mistralai-index.md) | AI search index | Mistral search | Yes | Your content is not indexed for Mistral search. |
| [MistralAI-Training](https://getrankroot.com/ai-bots/mistralai-training.md) | AI training crawler | Mistral model training | Yes | Your content is not collected for Mistral's training datasets. |

## User agent string

```
Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; MistralAI-Training/1.0; +https://docs.mistral.ai/robots)
```

## How to verify it's really MistralAI-Training

Find its requests in your server logs:

```
grep -i "MistralAI-Training" access.log
```

Mistral publishes IP lists for MistralAI-User and MistralAI-Index.

Mistral AI does not publish an IP list or reverse DNS method for MistralAI-Training, so the user agent is the only signal.

## Common mistakes

- Writing a User-agent: MistralAI-Training group and expecting your User-agent: * rules to still apply. A crawler follows only the most specific group that matches it and ignores the others, so repeat any shared Disallow lines inside the MistralAI-Training group.
- Leaving a CDN or host "block AI bots" setting on and only checking robots.txt. Those settings block requests at the network edge. Cloudflare, for example, changed its defaults on September 15, 2026 for customers who did not opt out: bots it classifies as Training or Agent are blocked on pages that display ads, and multi-purpose crawlers that combine Search and Training are affected by the Training block.

## FAQ

### What is MistralAI-Training?

Builds datasets for training Mistral's generative models. Not used for search indexing or live answers.

### Does MistralAI-Training respect robots.txt?

Yes, according to Mistral AI.

### What happens if I block MistralAI-Training?

Your content is not collected for Mistral's training datasets.

### Should I block MistralAI-Training?

Your choice.

## Sources

- [Official documentation from Mistral AI](https://docs.mistral.ai/robots)

Check whether MistralAI-Training can reach a site: https://getrankroot.com/tools/ai-crawler-checker or `GET https://getrankroot.com/api/bots?url=example.com`. Full list: https://getrankroot.com/ai-bots.md