README.md

---
library_name: transformers
tags:
- llama-factory
license: mit
datasets:
- OALL/details_unsloth__llama-3-8b-bnb-4bit
language:
- zh
---

# 將模型下載到Ollama當中使用

> [!NOTE]
> Ollama只支援`GGUF`檔喔，請先參考我的github，產生`量化模型檔`

## 請先安裝huggingface-hub
```python
pip install huggingface-hub
```

# 登入huggingface-cli

```
huggingface-cli login
```
![image/png](https://cdn-uploads.huggingface.co/production/uploads/66c1f6ca2777c0507076470b/tK8RTOZ_2rXFq8zFRiroV.png)


# 上傳單個檔案(量化模型)

```
huggingface-cli upload markhsu0818/LLamaTrainModel Llama3-8B-Chinese-Chat-Q4-terry-new.gguf ^
https://huggingface.co/markhsu0818/LLamaTrainModel/
```

# 下載單個檔案(量化模型)

```
huggingface-cli download markhsu0818/LLamaTrainModel ^
Llama3-8B-Chinese-Chat-Q4-terry-new.gguf ^
--local-dir downloads ^
--local-dir-use-symlinks False
```

![image/png](https://cdn-uploads.huggingface.co/production/uploads/66c1f6ca2777c0507076470b/Cj3WR7OqKvyaUosvxgaF8.png)


# 下載整個專案

```
Huggingface-cli download markhsu0818/LLamaTrainModel
```


![image/png](https://cdn-uploads.huggingface.co/production/uploads/66c1f6ca2777c0507076470b/ez8s1e63q8GNel76jEc0o.png)

# 參考連結

[將 Hugging Face 上的模型安裝到 OLlama](https://hackmd.io/@flagmaker/HkQHhlYyA?utm_source=preview-mode&utm_medium=rec)


<!-- Provide a quick summary of what the model is/does. -->