Dalian University of Technology's VA-Bench: Top Multimodal Models Finish Only Half of Robot Tasks

T2Covered by 1 source · first reported by China: SCMP Tech / TechNode / Pandaily / KrASIA / Sixth Tone at 05 Oct, 07:14 UTC

VA-Bench from Dalian University of Technology tests 12 multimodal model setups on robot-arm tasks; Qwen3.8-max leads at 53.93%, and none finishes a strict long-horizon task.

hardware

Sources

• T2Dalian University of Technology's VA-Bench: Top Multimodal Models Finish Only Half of Robot TasksChina: SCMP Tech / TechNode / Pandaily / KrASIA / Sixth Tone05 Oct, 07:14