Mistral

标签	日期	备注
`v0.3` `最新`	05/22/2024	Mistral 7B 的新版本，支持函数调用。
`v0.2`	03/23/2024	Mistral 7B 的一个小版本更新
`v0.1`	09/27/2023	初始版本

函数调用

Mistral 0.3 支持使用 Ollama 的**原始模式**进行函数调用。

_{原始提示示例}

[AVAILABLE_TOOLS] [{"type": "function", "function": {"name": "get_current_weather", "description": "Get the current weather", "parameters": {"type": "object", "properties": {"location": {"type": "string", "description": "The city and state, e.g. San Francisco, CA"}, "format": {"type": "string", "enum": ["celsius", "fahrenheit"], "description": "The temperature unit to use. Infer this from the users location."}}, "required": ["location", "format"]}}}][/AVAILABLE_TOOLS][INST] What is the weather like today in San Francisco [/INST]

_响应示例

[TOOL_CALLS] [{"name": "get_current_weather", "arguments": {"location": "San Francisco, CA", "format": "celsius"}}]

有关原始模式的更多信息，请参阅API 文档。

变体


`指令（instruct）`	指令模型会遵循指令
`文本（text）`	文本模型是基础模型，未经过任何对话方面的微调，最适合用于简单的文本完成。

使用

CLI

指令（instruct）

ollama run mistral

API

示例

curl -X POST https://127.0.0.1:11434/api/generate -d '{
  "model": "mistral",
  "prompt":"Here is a story about llamas eating grass"
 }'

参考资料

HuggingFace

Mistral AI 新闻稿

Mistral is a 7B parameter model, distributed with the Apache license. It is available in both instruct (instruction following) and text completion.

The Mistral AI team has noted that Mistral 7B:

- Outperforms Llama 2 13B on all benchmarks
- Outperforms Llama 1 34B on many benchmarks
- Approaches CodeLlama 7B performance on code, while remaining good at English tasks

### Versions

| Tag             | Date       | Notes                                                       |
| --------------- | ---------- | ----------------------------------------------------------- |
| `v0.3` `latest` | 05/22/2024 | A new version of Mistral 7B that supports function calling. |
| `v0.2`          | 03/23/2024 | A minor release of Mistral 7B                           |
| `v0.1`          | 09/27/2023 | Initial release                                             |

### Function calling

Mistral 0.3 supports function calling with Ollama's **raw mode**.

<sub>Example raw prompt</sub>
```
[AVAILABLE_TOOLS] [{"type": "function", "function": {"name": "get_current_weather", "description": "Get the current weather", "parameters": {"type": "object", "properties": {"location": {"type": "string", "description": "The city and state, e.g. San Francisco, CA"}, "format": {"type": "string", "enum": ["celsius", "fahrenheit"], "description": "The temperature unit to use. Infer this from the users location."}}, "required": ["location", "format"]}}}][/AVAILABLE_TOOLS][INST] What is the weather like today in San Francisco [/INST]
```

<sub>Example response</sub>

```
[TOOL_CALLS] [{"name": "get_current_weather", "arguments": {"location": "San Francisco, CA", "format": "celsius"}}]
```

For more information on raw mode, see the [API documentation](https://github.com/ollama/ollama/blob/main/docs/api.md#request-raw-mode).

### Variations

|            |                                                                                                                                    |
| ---------- | ---------------------------------------------------------------------------------------------------------------------------------- |
| `instruct` | Instruct models follow instructions                                                                                                |
| `text`     | Text models are the base foundation model without any fine-tuning for conversations, and are best used for simple text completion. |

## Usage

### CLI

Instruct:

```
ollama run mistral
```

### API

Example:

```
curl -X POST https://127.0.0.1:11434/api/generate -d '{
  "model": "mistral",
  "prompt":"Here is a story about llamas eating grass"
 }'
```

## References

[HuggingFace](https://hugging-face.cn/mistralai/Mistral-7B-Instruct-v0.2)

[Mistral AI News Release](https://mistral.ai/news/announcing-mistral-7b/)

粘贴、拖放或点击上传图像（.png、.jpeg、.jpg、.svg、.gif）