hfl
/

File size: 870 Bytes
47bc728
 
d45a7b9
 
 
47bc728
cc7fe76
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
26
27
28
29
30
31
32
33
34
35
36
37
---
license: apache-2.0
language:
- zh
- en
---

# Chinese-Mixtral-Instruct-GGUF

This repository contains the GGUF-v3 models (llama.cpp compatible) for **Chinese-Mixtral-Instruct** (chat/instruction model).

## Performance

Metric: PPL, lower is better

| Quant | PPL  |
| ----- | ---- |
| IQ2_XXS | - |
| IQ2_XS | - |
| Q2_K  | -    |
| Q3_K  | -     |
| Q4_0  | -      |
| Q4_K  | -    |
| Q5_0  | -     |
| Q5_K  | -    |
| Q6_K  | -     |
| Q8_0  | -     |
| F16   |   x   |

Due to the file size limitation, for F16 model, please use `cat` command to concatenate all parts into a single file. **You must concatenate these parts in order.**


## Others

For Hugging Face version, please see: https://huggingface.co/hfl/chinese-mixtral-instruct

Please refer to [https://github.com/ymcui/Chinese-Mixtral/](https://github.com/ymcui/Chinese-Mixtral/) for more details.