Hy-MT

中文&nbsp ｜ English

 Hugging Face  |  
 ModelScope  |  

️ Official Website  |  
️ Demo    

Github

NOTICE:

We have released the Hy-MT2 series of translation models, offering improved performance and excellent instruction-following capabilities. The link to the new model collection is: https://huggingface.co/collections/tencent/hy-mt2

We are excited to announce our official partnership with WMT26. We welcome all participants to use our HY-MT model during the competition. Teams that use HY-MT and achieve notable results will be eligible for cash prizes. For more details, please contact us at [email protected].

To help you get started with HY-MT training more quickly, we have provided a Training Tutorial. You can access it via the link.

Model Introduction

Hunyuan Translation Model Version 1.5 includes a 1.8B translation model, HY-MT1.5-1.8B, and a 7B translation model, HY-MT1.5-7B. Both models focus on supporting mutual translation across 33 languages and incorporating 5 ethnic and dialect variations. Among them, HY-MT1.5-7B is an upgraded version of our WMT25 championship model, optimized for explanatory translation and mixed-language scenarios, with newly added support for terminology intervention, contextual translation, and formatted translation. Despite having less than one-third the parameters of HY-MT1.5-7B, HY-MT1.5-1.8B delivers translation performance comparable to its larger counterpart, achieving both high speed and high quality. After quantization, the 1.8B model can be deployed on edge devices and support real-time translation scenarios, making it widely applicable.

Key Features and Advantages

• HY-MT1.5-1.8B achieves the industry-leading performance among models of the same size, surpassing most commercial translation APIs.
• HY-MT1.5-1.8B supports deployment on edge devices and real-time translation scenarios, offering broad applicability.
• HY-MT1.5-7B, compared to its September open-source version, has been optimized for annotated and mixed-language scenarios.
• Both models support terminology intervention, contextual translation, and formatted translation.

Related News

• 2025.12.30, we have open-sourced HY-MT1.5-1.8B and HY-MT1.5-7B on Hugging Face.
• 2025.9.1, we have open-sourced Hunyuan-MT-7B , Hunyuan-MT-Chimera-7B on Hugging Face.

Performance

You can refer to our technical report for more experimental results and analysis.

Technical Report

 

Model Links

Model Name	Description	Download
HY-MT1.5-1.8B	Hunyuan 1.8B translation model	Model
HY-MT1.5-1.8B-FP8	Hunyuan 1.8B translation model, fp8 quant	Model
HY-MT1.5-1.8B-GPTQ-Int4	Hunyuan 1.8B translation model, int4 quant	Model
HY-MT1.5-1.8B-GGUF	Hunyuan 1.8B translation model, llama.cpp	Model
HY-MT1.5-7B	Hunyuan 7B translation model	Model
HY-MT1.5-7B-FP8	Hunyuan 7B translation model, fp8 quant	Model
HY-MT1.5-7B-GGUF	Hunyuan 7B translation model, llama.cpp	Model

Prompts

Note: The following source_language and target_language should both use the full names of the languages; use the full Chinese names for Chinese instruction and the full English names for English instruction.

Prompt Template for ZHXX Translation.

---

将以下文本翻译为{target_language}，注意只需要输出翻译后的结果，不要额外解释：

{source_text}

---

Prompt Template for XXXX Translation, excluding ZHXX.

---

Translate the following segment into {target_language}, without additional explanation.

{source_text}

---

Prompt Template for terminology intervention.

---

参考下面的翻译：
{source_term} 翻译成 {target_term}

将以下文本翻译为{target_language}，注意只需要输出翻译后的结果，不要额外解释：
{source_text}

---

Prompt Template for contextual translation.

---

{context}
参考上面的信息，把下面的文本翻译成{target_language}，注意不需要翻译上文，也不要额外解释：
{source_text}

---

Prompt Template for formatted translation.

---

将以下之间的文本翻译为中文，注意只需要输出翻译后的结果，不要额外解释，原文中的标签表示标签内文本包含格式信息，需要在译文中相应的位置尽量保留该标签。输出格式为：str

{src_text_with_format}

---

 

Use with transformers

First, please install transformers, recommends v4.56.0

pip install transformers==4.56.0

!!! 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.

The following code snippet shows how to use the transformers library to load and apply the model.

we use tencent/HY-MT1.5-1.8B for example

…

We recommend using the following set of parameters for inference. Note that our model does not have the default system_prompt.

{
  "top_k": 20,
  "top_p": 0.6,
  "repetition_penalty": 1.05,
  "temperature": 0.7
}

 

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

Training Data Format

If you need to fine-tune our Instruct model, we recommend processing the data into the following format.

…

 

Train with LLaMA-Factory

In the following chapter, we will introduce how to use LLaMA-Factory to fine-tune the Hunyuan model.

Prerequisites

Verify installation of the following dependencies:

• LLaMA-Factory: Follow official installation guide
• DeepSpeed (optional): Follow official installation guide
• Transformer Library: Use the companion branch (Hunyuan-submitted code is pending review)

  pip install git+https://github.com/huggingface/transformers@4970b23cedaf745f963779b4eae68da281e8c6ca
  

Data preparation

We need to prepare a custom dataset:

1. Organize your data in json format and place it in the data directory in LLaMA-Factory. The current implementation uses the sharegpt dataset format, which requires the following structure:

[
  {
    "messages": [
      {
        "role": "system",
        "content": "System prompt (optional)"
      },
      {
        "role": "user",
        "content": "Human instruction"
      },
      {
        "role": "assistant",
        "content": "Model response"
      }
    ]
  }
]

Refer to the Data Format section mentioned earlier for details.

1. Define your dataset in the data/dataset_info.json file using the following format:

"dataset_name": {
  "file_name": "dataset.json",
  "formatting": "sharegpt",
  "columns": {
    "messages": "messages"
  },
  "tags": {
    "role_tag": "role",
    "content_tag": "content",
    "user_tag": "user",
    "assistant_tag": "assistant",
    "system_tag": "system"
  }
}

Training execution

1. Copy all files from the llama_factory_support/example_configs directory to the example/hunyuan directory in LLaMA-Factory.
2. Modify the model path and dataset name in the configuration file hunyuan_full.yaml. Adjust other configurations as needed:

### model
model_name_or_path: [!!!add the model path here!!!]

### dataset
dataset: [!!!add the dataset name here!!!]

1. Execute training commands: *​​Single-node training​​ Note: Set the environment variable DISABLE_VERSION_CHECK to 1 to avoid version conflicts.

   export DISABLE_VERSION_CHECK=1
   llamafactory-cli train examples/hunyuan/hunyuan_full.yaml
   

   *Multi-node training​​ Execute the following command on each node. Configure NNODES, NODE_RANK, MASTER_ADDR, and MASTER_PORT according to your environment:

   export DISABLE_VERSION_CHECK=1
   FORCE_TORCHRUN=1 NNODES=${NNODES} NODE_RANK=${NODE_RANK} MASTER_ADDR=${MASTER_ADDR} MASTER_PORT=${MASTER_PORT} \
   llamafactory-cli train examples/hunyuan/hunyuan_full.yaml
   

 

Quantization Compression

We used our own AngelSlim compression tool to produce FP8 and INT4 quantization models. AngelSlim is a toolset dedicated to creating a more user-friendly, comprehensive and efficient model compression solution.

FP8 Quantization

We use FP8-static quantization, FP8 quantization adopts 8-bit floating point format, through a small amount of calibration data (without training) to pre-determine the quantization scale, the model weights and activation values will be converted to FP8 format, to improve the inference efficiency and reduce the deployment threshold. We you can use AngelSlim quantization, you can also directly download our quantization completed open source model to use AngelSlim.

Deployment

For deployment, you can use frameworks such as TensorRT-LLM, vLLM, or SGLang to serve the model and create an OpenAI-compatible API endpoint.

image: https://hub.docker.com/r/hunyuaninfer/hunyuan-7B/tags

TensorRT-LLM

Docker Image

We provide a pre-built Docker image based on the latest version of TensorRT-LLM.

We use tencent/Hunyuan-7B-MT for example

• To get started:

docker pull docker.cnb.cool/tencent/hunyuan/hunyuan-7b:hunyuan-7b-trtllm

docker run --privileged --user root --name hunyuanLLM_infer --rm -it --ipc=host --ulimit memlock=-1 --ulimit stack=67108864 --gpus=all hunyuaninfer/hunyuan-7b:hunyuan-7b-trtllm

• Prepare Configuration file:

cat >/path/to/extra-llm-api-config.yml &1 | tee log_server.txt

• After running service script successfully, run the request script

curl http://0.0.0.0:8000/v1/chat/completions -H 'Content-Type: application/json' -d '{
"model": "hunyuan",
"messages": [
    {
        "role": "system",
        "content": [{"type": "text", "text": "You are a