# macos 跑EXO

## A.什麼是EXO？

EXO 在X算是熱門，就是它可以將macOS/linux Cluster 跑大模型，而且是十分簡單就可以實現，跑出來的效果還相當不錯，幾個重點是：

1.使用mac min 暫用空間小

2.花費電力低，溫度也低

3.連接設備簡單，thunderbolt 完成

4.價格不貴，貨源多，一台mac mini 不到4k，已經可以跑.15b，cluster 幾台已經很快

## B. 安裝 miniconda

由於exo 需要python 3.12 以上，而且需要很多東西，個人測試後最簡單還是用conda，方法如下，粘貼到終端機執行就可以

```
mkdir -p ~/miniconda3
```

```
curl https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-arm64.sh -o ~/miniconda3/miniconda.sh
```

```
bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3
```

```
rm ~/miniconda3/miniconda.sh
```

安裝完成後開啟一個新的終端機（其實沒開也是可以的）

```
conda init --all
```

這樣miniconda 就完成安裝了，接下來要安裝MLX

## C.安裝MLX

要求macOS 是13.5以上

```
pip install mlx
```

```
conda install conda-forge::mlx
```

## D.安裝git

直接使用conda 安裝git就可以，和pip 一樣

`conda install git`

## E.安裝EXO

git clone <https://github.com/exo-explore/exo.git&#x20>;

cd exo&#x20;

pip install -e .

source install.sh

## F.執行EXO

exo


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://rockfu.gitbook.io/rockdocs/ai-ri-ji/macos-pao-exo.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.