File size: 1,763 Bytes
bfe126a e975839 5044120 bfe126a c210776 1e7e62d 1fda25a 1e7e62d c210776 1fda25a c210776 a69f340 a994d33 730e86f 2190f8a 9a68d23 2190f8a c210776 68e6d2e c210776 41b99b8 |
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 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 |
---
license: apache-2.0
language:
- zh
- en
tags:
- moe
---
# Chinese-Mixtral-GGUF
<p align="center">
<a href="https://github.com/ymcui/Chinese-Mixtral"><img src="https://ymcui.com/images/chinese-mixtral-banner.png" width="600"/></a>
</p>
**Chinese Mixtral GitHub repository: https://github.com/ymcui/Chinese-Mixtral**
This repository contains the GGUF-v3 models (llama.cpp compatible) for **Chinese-Mixtral** (this is not a chat/instruction model).
## Performance
Metric: PPL, lower is better
| Quant | PPL |
| ----- | ---- |
| IQ1_S | 20.7314 +/- 0.22627 |
| IQ2_XXS | 8.5981 +/- 0.09267 |
| IQ2_XS | 6.9784 +/- 0.07476 |
| Q2_K | 5.1846 +/- 0.05533 |
| IQ3_XXS | 4.5990 +/- 0.04969 |
| Q3_K | 4.5545 +/- 0.04893 |
| Q4_0 | 4.4917 +/- 0.04844 |
| Q4_K | 4.4488 +/- 0.04813 |
| Q5_0 | 4.4224 +/- 0.04753 |
| Q5_K | 4.4192 +/- 0.04768 |
| Q6_K | 4.4092 +/- 0.04758 |
| Q8_0 | 4.4076 +/- 0.04746 |
| 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
- If you have questions/issues regarding this model, please submit an issue through https://github.com/ymcui/Chinese-Mixtral/.
## Citation
Please consider cite our paper if you use the resource of this repository.
Paper link: https://arxiv.org/abs/2403.01851
```
@article{chinese-mixtral,
title={Rethinking LLM Language Adaptation: A Case Study on Chinese Mixtral},
author={Cui, Yiming and Yao, Xin},
journal={arXiv preprint arXiv:2403.01851},
url={https://arxiv.org/abs/2403.01851},
year={2024}
}
``` |