File size: 620 Bytes
8a6e3f3
 
79892e2
 
 
8a6e3f3
79892e2
 
 
65572df
 
d54e538
 
 
 
 
 
 
79892e2
 
 
 
65572df
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
---
license: apache-2.0
language:
- zh
- en
---

# gpt4-x-vicuna-13B-chinese

使用最新的优质模型gpt4-x-vicuna-13B + 中文Lora 合并而来  
苦于没有好的中文模型,自己摸索搞定了,现在中文效果还行,欢迎试用反馈  

# 已知问题

- uncensored模型会变得保守


# 权重来源

- Lora权重来源: https://github.com/ymcui/Chinese-LLaMA-Alpaca 
- gpt4-x-vicuna-13B量化前的[原模型](https://huggingface.co/NousResearch/gpt4-x-vicuna-13b/tree/main)

# 使用方法

Q4 ggml格式,请用llama.cpp或者封装好的LocalAi或者text generation ui来加载运行模型