---
license: other
language:
- zh
- en
datasets:
- thomas-yanxin/MT-SFT-ShareGPT
---


The main purpose of this model is to validate the usability of [thomas-yanxin/MT-SFT-ShareGPT](https://huggingface.co/datasets/thomas-yanxin/MT-SFT-ShareGPT), i.e., the quality of the data is all you need. We found that when we meticulously extract the data through a better data governance approach, the corresponding model results can be vastly improved, even if only through SFT.

Here are the results from our OpenCompass evaluation：

| Classification | Benchmarks | Models |
| :------------: | :--------: | :--------: |
|        | 名称   | XinYuan-Qwen2-7B |
| English       | MMLU   | 68.71            |
|            | MMLU-Pro | 30.56            |
|            | Theorem QA | 25.3             |
|            | GPQA    | 29.2             |
|            | BBH     | 60.3             |
|        | IFEval (Prompt Strict-Acc.) | 39.2 |
|        | ARC-C   | 87.5             |
| Math       | GSM8K   | 75.4             |
|            | MATH    | 34.76            |
| Chinese       | C-EVAL  | 82.0             |
|            | CMMLU   | 77.9             |
| Code       | MBPP    | 50.6             |
|            | HumanEval | 70.1             |