Tencent releases its large reasoning model, Hunyuan-T1. T1 has achieved high rankings on several key artificial intelligence benchmarks. This positions it among the leading models worldwide.

Recent evaluations indicate that Hunyuan-T1 scored 87.2 on MMLU-Pro, a dataset assessing foundational reasoning abilities in large language models. This score places it second only to OpenAI o1.

Moreover, Hunyuan T1 performed well on public benchmarks such as CEval, AIME, and Zebra Logic. These benchmarks measure general knowledge and reasoning in both Chinese and English.

Recent Tencent developments