Tencent Hunyuan releases Hy4, ranking among the strongest open-source models.

On November 4, Tencent Hunyuan released Hy4, its latest-generation large language model. Compared with the previous Hy4 preview, the new model further expands the scale of pre-training and post-training, introduces visual understanding capabilities for the first time, and places emphasis on strengthening instruction following, long-text understanding, long-horizon Agent capabilities, and performance on real productivity tasks. Hy4 has a total parameter scale of 770B, adopts a mixture-of-experts architecture, and supports computer operation for the first time, enabling it to execute complex tasks by combining command-line and graphical interfaces. With another comprehensive leap in model capabilities, Hy has for the first time ranked among the strongest open-source models. Hy4 will soon open-source its model weights, release a model system card, and actively advance more responsible open-sourcing, with a focus on risks such as cyber and biological security.