---
title: Hunyuan-MT-7B
canonical_url: "https://www.modelscope.cn/models/Tencent-Hunyuan/Hunyuan-MT-7B"
md_url: "https://www.modelscope.cn/models/Tencent-Hunyuan/Hunyuan-MT-7B.md"
repository: Tencent-Hunyuan/Hunyuan-MT-7B
chinese_name: Hunyuan-MT-7B
last_updated: 2025-09-03
pipeline_tag: translation
tasks:
  - translation
model_type:
  - hunyuan_v1_dense
architectures:
  - HunYuanDenseV1ForCausalLM
parameters: 8.0B
tensor_type:
  - BF16
library_name:
  - transformer
  - safetensors
  - pytorch
frameworks:
  - Pytorch
inference_backends:
  - "deploy_task text"
  - "sglang 0.5.2"
downloads: 22596
stars: 63
---

# Hunyuan-MT-7B

> Hunyuan-MT-7B - Tencent-Hunyuan 在 ModelScope 开源的模型。🤗  Hugging Face   |     ModelScope   |  

Tencent-Hunyuan/Hunyuan-MT-7B 是 ModelScope 魔搭社区上的 8.0B 参数translation模型，可用 deploy_task text、sglang 0.5.2 部署。

- **Repository**: Tencent-Hunyuan/Hunyuan-MT-7B
- **Tasks**: translation
- **Parameters**: 8.0B
- **Inference backends**: deploy_task text, sglang 0.5.2
- **Downloads**: 22596
- **Stars**: 63
- **Last updated**: 2025-09-03

Source: https://www.modelscope.cn/models/Tencent-Hunyuan/Hunyuan-MT-7B

---

<p align="center">
 <img src="https://dscache.tencent-cloud.cn/upload/uploader/hunyuan-64b418fd052c033b228e04bc77bbc4b54fd7f5bc.png" width="400"/> <br>
</p><p></p>


<p align="center">
    🤗&nbsp;<a href="https://huggingface.co/collections/tencent/hunyuan-mt-68b42f76d473f82798882597"><b>Hugging Face</b></a>&nbsp;&nbsp;|&nbsp;&nbsp;
    <img src="https://avatars.githubusercontent.com/u/109945100?s=200&v=4" width="16"/>&nbsp;<a href="https://modelscope.cn/collections/Hunyuan-MT-2ca6b8e1b4934f"><b>ModelScope</b></a>&nbsp;&nbsp;|&nbsp;&nbsp;
</p>

<p align="center">
    🖥️&nbsp;<a href="https://hunyuan.tencent.com" style="color: red;"><b>Official Website</b></a>&nbsp;&nbsp;|&nbsp;&nbsp;
    🕹️&nbsp;<a href="https://hunyuan.tencent.com/modelSquare/home/list"><b>Demo</b></a>&nbsp;&nbsp;&nbsp;&nbsp;
</p>

<p align="center">
    <a href="https://github.com/Tencent-Hunyuan/Hunyuan-MT"><b>GITHUB</b></a>
</p>




## 模型介绍

混元翻译模型，包含一个翻译模型Hunyuan-MT-7B和一个集成模型Hunyuan-MT-Chimera。翻译模型用来将待翻译的文本翻译成目标语言，集成模型用来把翻译模型的多个翻译结果集成为一个更好的翻译。重点支持33语种互译，支持5种民汉语言。



### 核心特性与优势
- WMT25参赛31语种之中30语种获得第一名的成绩。
- Hunyuan-MT-7B尺寸业界效果最优
- Hunyuan-MT-Chimera-7B是业界首个开源翻译集成模型，可以将翻译效果再拉高一个档次
- 提出了一个完整的翻译模型训练范式，从Pretrain->CPT->SFT->翻译强化->集成强化，翻译效果达到同尺寸SOTA。


## 新闻
<br>

* 2025.9.1 我们在Hugging Face开源了 **Hunyuan-MT-7B** , **Hunyuan-MT-Chimera-7B**。


## 模型链接
| Model Name  | Description | Download |
| ----------- | ----------- |-----------
| Hunyuan-MT-7B  | 混元7B翻译模型 |🤗 [Model](https://huggingface.co/tencent/Hunyuan-MT-7B)|
| Hunyuan-MT-7B-fp8 | 混元7B翻译模型，fp8量化    | 🤗 [Model](https://huggingface.co/tencent/Hunyuan-MT-7B-fp8)|
| Hunyuan-MT-Chimera | 混元7B翻译集成模型    | 🤗 [Model](https://huggingface.co/tencent/Hunyuan-MT-Chimera-7B)|
| Hunyuan-MT-Chimera-fp8 | 混元7B翻译集成模型，fp8量化     | 🤗 [Model](https://huggingface.co/tencent/Hunyuan-MT-Chimera-7B-fp8)|


## Prompts

### Prompt Template for ZH<=>XX Translation.
---

把下面的文本翻译成`<target_language>`，不要额外解释。

`<source_text>`

---

### Prompt Template for XX<=>XX Translation, excluding ZH<=>XX.
---

Translate the following segment into `<target_language>`, without additional explanation.

`<source_text>`

---

### Prompt Template for Hunyuan-MT-Chmeria-7B

---

Analyze the following multiple `<target_language>` translations of the `<source_language>` segment surrounded in triple backticks and generate a single refined `<target_language>` translation. Only output the refined translation, do not explain.

The `<source_language>` segment:
```<source_text>```

The multiple `<target_language>` translations:
1. ```<translated_text1>```
2. ```<translated_text2>```
3. ```<translated_text3>```
4. ```<translated_text4>```
5. ```<translated_text5>```
6. ```<translated_text6>```

---

## 使用 transformers 推理
首先，需要安装最新版本的transformers，推荐v4.56.0
```SHELL
pip install transformers==4.56.0
```

以下代码片段展示了如何使用 transformers 库加载和使用模型。

*!!! If you want to load fp8 model with transformers, you need to change the name"ignored_layers" in config.json to "ignore" and upgrade the compressed-tensors to compressed-tensors-0.11.0.*


```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import os

model_name_or_path = "tencent/Hunyuan-MT-7B"

tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
model = AutoModelForCausalLM.from_pretrained(model_name_or_path, device_map="auto")  # You may want to use bfloat16 and/or move to GPU here
messages = [
    {"role": "user", "content": "Translate the following segment into Chinese, without additional explanation.\n\nGet something off your chest"},
]
tokenized_chat = tokenizer.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=False,
    return_tensors="pt"
)

outputs = model.generate(tokenized_chat.to(model.device), max_new_tokens=2048)
output_text = tokenizer.decode(outputs[0])
```

Supported languages:
| Languages         | Abbr.   | Chinese Names   |
|-------------------|---------|-----------------|
| Chinese           | zh      | 中文            |
| English           | en      | 英语            |
| French            | fr      | 法语            |
| Portuguese        | pt      | 葡萄牙语        |
| Spanish           | es      | 西班牙语        |
| Japanese          | ja      | 日语            |
| Turkish           | tr      | 土耳其语        |
| Russian           | ru      | 俄语            |
| Arabic            | ar      | 阿拉伯语        |
| Korean            | ko      | 韩语            |
| Thai              | th      | 泰语            |
| Italian           | it      | 意大利语        |
| German            | de      | 德语            |
| Vietnamese        | vi      | 越南语          |
| Malay             | ms      | 马来语          |
| Indonesian        | id      | 印尼语          |
| Filipino          | tl      | 菲律宾语        |
| Hindi             | hi      | 印地语          |
| Traditional Chinese | zh-Hant| 繁体中文        |
| Polish            | pl      | 波兰语          |
| Czech             | cs      | 捷克语          |
| Dutch             | nl      | 荷兰语          |
| Khmer             | km      | 高棉语          |
| Burmese           | my      | 缅甸语          |
| Persian           | fa      | 波斯语          |
| Gujarati          | gu      | 古吉拉特语      |
| Urdu              | ur      | 乌尔都语        |
| Telugu            | te      | 泰卢固语        |
| Marathi           | mr      | 马拉地语        |
| Hebrew            | he      | 希伯来语        |
| Bengali           | bn      | 孟加拉语        |
| Tamil             | ta      | 泰米尔语        |
| Ukrainian         | uk      | 乌克兰语        |
| Tibetan           | bo      | 藏语            |
| Kazakh            | kk      | 哈萨克语        |
| Mongolian         | mn      | 蒙古语          |
| Uyghur            | ug      | 维吾尔语        |
| Cantonese         | yue     | 粤语            |


Citing Hunyuan-MT:

```bibtex
@misc{hunyuanmt2025,
  title={Hunyuan-MT Technical Report},
  author={Mao Zheng, Zheng Li, Bingxin Qu, Mingyang Song, Yang Du, Mingrui Sun, Di Wang, Tao Chen, Jiaqi Zhu, Xingwu Sun, Yufei Wang, Can Xu, Chen Li, Kai Wang, Decheng Wu},
  howpublished={\url{https://github.com/Tencent-Hunyuan/Hunyuan-MT}},
  year={2025}
}
```