This is a pretrained FlowerVLA model for robotic manipulation trained on the LIBERO Spatial dataset. Flower is an efficient Vision-Language-Action Flow policy for robot learning that only contains 1B parameters.

FlowerVLA is a novel architecture that:

This checkpoint contains weights for the LIBERO Spatial challenge and achieves these results:

avg_seq_len success rate 0.9681089520454407 pick_up_the_black_bowl_between_the_plate_and_the_ramekin_and_place_it_on_the_plate with success 0.9791666666666666 pick_up_the_black_bowl_next_to_the_ramekin_and_place_it_on_the_plate with success 0.9807692307692308 pick_up_the_black_bowl_from_table_center_and_place_it_on_the_plate with success 0.9807692307692308 pick_up_the_black_bowl_on_the_cookie_box_and_place_it_on_the_plate with success 1.0 pick_up_the_black_bowl_in_the_top_drawer_of_the_wooden_cabinet_and_place_it_on_the_plate with success 1.0 pick_up_the_black_bowl_on_the_ramekin_and_place_it_on_the_plate with success 0.8621794871794872 pick_up_the_black_bowl_next_to_the_cookie_box_and_place_it_on_the_plate with success 1.0 pick_up_the_black_bowl_on_the_stove_and_place_it_on_the_plate with success 1.0 pick_up_the_black_bowl_next_to_the_plate_and_place_it_on_the_plate with success 0.9166666666666666 pick_up_the_black_bowl_on_the_wooden_cabinet_and_place_it_on_the_plate with success 0.9615384615384616

Check out our full model implementation on Github todo and follow the instructions in the readme to test the model on one of the environments.

@inproceedings{ reuss2025flower, # Add citation when available }

This model is released under the MIT license.