# Mistral Nemo 12B

**Summary:** Mistral Nemo is a compact yet powerful 12-billion parameter language model co-developed with NVIDIA, designed specifically for conversational agents and virtual assistants. This French-engineered model features an advanced Tekken tokenizer that provides superior efficiency for European languages, making it ideal for applications requiring fast response times while maintaining high-quality multilingual performance across German, French, Spanish, Italian, and Portuguese.

|                                                                                                     **Intelligence**                                                                                                     |                                                                                                                                                                                                                                                                                                                       **Speed**                                                                                                                                                                                                                                                                                                                       |                                                                                                                                                                                                                 **Sovereignty**                                                                                                                                                                                                                 |                                                                                                                                                                                                                                                                                                                       **Input**                                                                                                                                                                                                                                                                                                                      |                                                                                                                                                                                                                                                                                                                      **Output**                                                                                                                                                                                                                                                                                                                      |
| :----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: | :---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: | :---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
| ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-b23196ddc0cba1be0b981aa5572379cec1538be3%2Fai-model-hub-intelligence.png?alt=media) | ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-be3201cb2eba83650220699adf5b3d9120c83377%2Fai-model-hub-speed.png?alt=media) ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-be3201cb2eba83650220699adf5b3d9120c83377%2Fai-model-hub-speed.png?alt=media) ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-be3201cb2eba83650220699adf5b3d9120c83377%2Fai-model-hub-speed.png?alt=media) | ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-2c04b225a16490c4ff8c3e062bb166f25e05e1c2%2Fai-model-hub-sovereignty.png?alt=media) ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-2c04b225a16490c4ff8c3e062bb166f25e05e1c2%2Fai-model-hub-sovereignty.png?alt=media) | ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-cc6707e286bceb4641047e45e095950e8db880fd%2Fai-model-hub-text.png?alt=media) ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-bac2752d06f18e86dc7f0b9531ab32ec58f30aec%2Fai-model-hub-image.png?alt=media) ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-8b0332538fcac6644893a504f9fbbd1ba2b56d21%2Fai-model-hub-audio.png?alt=media) | ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-cc6707e286bceb4641047e45e095950e8db880fd%2Fai-model-hub-text.png?alt=media) ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-bac2752d06f18e86dc7f0b9531ab32ec58f30aec%2Fai-model-hub-image.png?alt=media) ![](https://1737632334-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MifAzdGvKLDTtvJP8sm%2Fuploads%2Fgit-blob-8b0332538fcac6644893a504f9fbbd1ba2b56d21%2Fai-model-hub-audio.png?alt=media) |
|                                                                                                           *Low*                                                                                                          |                                                                                                                                                                                                                                                                                                                         *High*                                                                                                                                                                                                                                                                                                                        |                                                                                                                                                                                                                    *Moderate*                                                                                                                                                                                                                   |                                                                                                                                                                                                                                                                                                                        *Text*                                                                                                                                                                                                                                                                                                                        |                                                                                                                                                                                                                                                                                                                        *Text*                                                                                                                                                                                                                                                                                                                        |

## Central parameters

**Description:** Mistral Nemo replaces Mistral 7B as the latest generation small language model, offering enhanced performance with 12.2B parameters and a 128k context window. The model utilizes fp8 quantization for optimized inference efficiency.

**Model identifier:** `mistralai/Mistral-Nemo-Instruct-2407`

## IONOS AI Model Hub Lifecycle and Alternatives

| **IONOS Launch** | **End of Life** |                                                                  **Alternative**                                                                 | **Successor** |
| :--------------: | :-------------: | :----------------------------------------------------------------------------------------------------------------------------------------------: | :-----------: |
|  *May 15, 2025*  |       N/A       | [<mark style="color:blue;">**Llama 3.1 (8B)**</mark>](https://docs.ionos.com/sections-test/guides/ai/ai-model-hub/models/llms/meta-llama-3-1-8b) |               |

## Origin

|                              **Provider**                              | **Country** |                                            **License**                                           | **Flavor** |   **Release**   |
| :--------------------------------------------------------------------: | :---------: | :----------------------------------------------------------------------------------------------: | :--------: | :-------------: |
| [<mark style="color:blue;">**Mistral AI**</mark>](https://mistral.ai/) |    France   | [<mark style="color:blue;">**License**</mark>](https://www.apache.org/licenses/LICENSE-2.0.html) |  Instruct  | *July 18, 2024* |

## Technology

| **Context window** | **Parameters** | **Quantization** | **Multilingual** |                                                **Further details**                                               |
| :----------------: | :------------: | :--------------: | :--------------: | :--------------------------------------------------------------------------------------------------------------: |
|       *128k*       |     *12.2B*    |       *fp8*      |       *Yes*      | [<mark style="color:blue;">**Hugging Face**</mark>](https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407) |

## Modalities

|     **Text**     |   **Image**   |   **Audio**   |
| :--------------: | :-----------: | :-----------: |
| Input and output | Not supported | Not supported |

## Endpoints

| **Chat Completions** | **Embeddings** | **Image generation** |
| :------------------: | :------------: | :------------------: |
|  v1/chat/completions |  Not supported |     Not supported    |

## Features

| **Streaming** | **Reasoning** | **Tool calling** |
| :-----------: | :-----------: | :--------------: |
|   Supported   | Not supported |     Supported    |

## Usage example

### Chat completions

The following example demonstrates how to use **Mistral Nemo** for instructional tasks, using a descriptive prompt to explain complex concepts simply.

**API Endpoint:** `POST https://openai.inference.de-txl.ionos.com/v1/chat/completions`

**Request:**

```json
{
  "model": "mistralai/Mistral-Nemo-Instruct-2407",
  "messages": [
    {
      "role": "user",
      "content": "Explain the concept of quantum entanglement to a 5-year-old using simple analogies."
    }
  ],
  "temperature": 0.7,
  "top_p": 0.9,
  "n": 1,
  "stream": false,
  "max_tokens": 1000
}
```

**Response:**

```json
{
  "id": "chatcmpl-456",
  "object": "chat.completion",
  "created": 1677652290,
  "model": "mistralai/Mistral-Nemo-Instruct-2407",
  "choices": [{
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "Imagine you have two magic dice. No matter how far apart they are—even if one is on Earth and the other is on Mars—if you roll a 6 on one, the other one will instantly show a 6 too! They are connected in a special way that lets them 'talk' to each other instantly."
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 25,
    "completion_tokens": 60,
    "total_tokens": 85
  }
}
```

## Rate limits

Rate limits ensure fair usage and reliable access to the AI Model Hub. In addition to the [<mark style="color:blue;">contract-wide rate limits</mark>](https://docs.ionos.com/sections-test/guides/ai/ai-model-hub/how-tos/rate-limits), no model-specific limits apply.