Recently, Tencent Hunyuan unveiled that its translation model, Hunyuan-MT-7B, has claimed the number-one spot on the Hugging Face model trending chart. Launched as an open-source model on September 1st, this lightweight translation model boasts 7 billion parameters and offers support for 33 languages, along with 5 types of mutual translations between ethnic languages/dialects and Chinese. In the International Conference on Computational Linguistics (COLING) WMT2025 competition, which wrapped up at the end of August, the model demonstrated its prowess by securing first place in 30 out of 31 language tests. These tests encompassed major languages like Chinese, English, and Japanese, as well as less commonly spoken languages such as Czech and Marathi. According to Tencent officials, when evaluated on the Flores200 dataset, the model not only substantially outperformed others of a similar size but also held its own against significantly larger models. At present, the model has been seamlessly integrated into services such as Tencent Meeting and WeCom, enhancing their multilingual capabilities.