RT-2: New model translates vision and language into actiondeepmind.com1 point·jw_mc··0 commentsOpen articleSaveView on HN