# Use Mistral for embeddings

**URL:** https://meta.discourse.org/t/use-mistral-for-embeddings/399854
**Category:** Feature
**Tags:** ai
**Created:** [April 3, 2026, 5:47am UTC](https://meta.discourse.org/t/use-mistral-for-embeddings/399854 "2026-04-03T05:47:49Z")
**Posts on this page:** 2
**Page:** 1

<div class="post-metadata">

### Author: ![RGJ](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/rgj/32/523185_2.png) [@RGJ](https://meta.discourse.org/u/RGJ)
#### Post date: [April 3, 2026, 5:47am UTC](https://meta.discourse.org/t/use-mistral-for-embeddings/399854/1 "2026-04-03T05:47:49Z")

</div>

We’ve ran into another issue using Mistral for embeddings.

- Per [this topic](https://meta.discourse.org/t/how-to-implement-mistral-with-embeddings/361095) use OpenAI as provider and the Mistral service URL as URL
- Select tokenizer, sequence length and distance function
- Set model name to ‘mistral-embed’

When a dimension is entered, Mistral cries about not supporting that

```plaintext
 Trying to contact the model returned this error: {{
   "object":"error",
   "message":{
      "detail":[
         {
            "type":"extra_forbidden",
            "loc":[
               "body",
               "dimensions"
            ],
            "msg":"Extra inputs are not permitted",
            "input":2000
         }
      ]
   },
   "type":"invalid_request_error",
   "param":null,
   "code":null,
   "raw_status_code":422
}

```

This is because [Mistral](https://docs.mistral.ai/api/endpoint/embeddings) calls this `output_dimension` so it is not completely OpenAI compatible.

When I leave out the `dimensions` parameter, “Run Test” works, but it also prevents me from saving the model, telling me that “dimensions” is a required parameter.

 ![image](https://global.discourse-cdn.com/meta/original/4X/3/e/d/3eda0e1d2304f06686b3287a32418c54949de732.png)

 ![image](https://global.discourse-cdn.com/meta/original/4X/c/6/f/c6fb836502dec6a03cf6c34bf223ee3acf82fc8a.png)

Being able to use Mistral is quite crucial for GDPR compliance, so it would be good if the dimensions parameter could be omitted (easy fix) or when Mistral could be a first-class provider (better).

---

<div class="post-metadata">

### Author: ![Falco](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/falco/32/179432_2.png) [@Falco](https://meta.discourse.org/u/Falco)
#### Post date: [April 6, 2026, 8:45pm UTC](https://meta.discourse.org/t/use-mistral-for-embeddings/399854/3 "2026-04-06T20:45:55Z")

</div>

It passes the test for me with this configuration, that I did following their documentation

 ![image](https://global.discourse-cdn.com/meta/original/4X/e/f/3/ef3c1fc5858077adb0e2161c19e552abfaa4fd9f.png)

That said, I’d recommend using a model that scores better, like Qwen 3 embedding model, and the myriad of fine-tunes from it.

> [@RGJ](#):
>
> Being able to use Mistral is quite crucial for GDPR compliance,

It ain’t the only GDPR compliant game in town, albeit it may be the first one that comes to people minds?

OpenRouter has a list [Models | OpenRouter](https://openrouter.ai/models?fmt=cards&output_modalities=embeddings&zdr=true), and people can also self-host their embeddings model, it is very doable, and much easier than self-hosting LLMs.