Hy-Embodied-0.5-VLA

Tencent's first Hunyuan vision-language-action model, from Tencent Robotics X and the Hunyuan team: a VLA and robot-learning stack for bimanual imitation learning, released with UMI-pretrained and RoboTwin-finetuned checkpoints plus the accompanying dataset. Apache-2.0. Marks Hunyuan's expansion from language/image/video/3D into embodied AI.