The first VTLA tactile model of Qianjue robot made its debut at the World Artificial Intelligence Conference.

The 2026 World Artificial Intelligence Conference will be held from July 17th to 20th. Qianjue Robot brought the VTLA physical touch model, visual-touch multimodal data set, core number acquisition equipment, and tactile sensor hardware matrix. Qianjue Robot founder Ma Daolin told reporters that, unlike traditional single-vision VLA models, the VTLA model innovatively integrates visual and tactile bimodal perception information, allowing robots to have both visual recognition capabilities and tactile perception capabilities. It can capture the physical properties of objects that cannot be obtained by a single visual device, and the recognition dimension and detection accuracy have been greatly upgraded.