mbreuss/flower_libero_goal

FlowerVLA - Vision-Language-Action Flow Model finetuned on LIBERO Goal

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

Model Description

FlowerVLA is a novel architecture that:

• Uses half of Florence-2 for multi-modal vision-language encoding
• Employs an novel transformer-based flow matching architecture
• Provides an efficient, versatile VLA policy with only ~1B parameters

Model Performance

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

avg_seq_len success rate 0.9665064215660095 open_the_middle_drawer_of_the_cabinet with success 0.9807692307692308 put_the_bowl_on_the_stove with success 1.0 put_the_wine_bottle_on_top_of_the_cabinet with success 0.9807692307692308 sr_open_the_top_drawer_and_put_the_bowl_inside with success 0.842948717948718 sr_put_the_bowl_on_top_of_the_cabinet with success 1.0 push_the_plate_to_the_front_of_the_stove with success 0.9407051282051282 put_the_cream_cheese_in_the_bowl with success 0.9807692307692308 turn_on_the_stove with success 1.0 put_the_bowl_on_the_plate with success 1.0 put_the_wine_bottle_on_the_rack with success 0.9391025641025641

Input/Output Specifications

Inputs

• RGB Static Camera: (B, T, 3, H, W) tensor
• RGB Gripper Camera: (B, T, 3, H, W) tensor
• Language Instructions: Text strings

Outputs

• Action Space: (B, T, 7) tensor representing delta EEF actions

Usage

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.

obs = {
    "rgb_obs": {
        "rgb_static": static_image,
        "rgb_gripper": gripper_image
    }
}
goal = {"lang_text": "pick up the blue cube"}
action = model.step(obs, goal)

Training Details

Configuration

• Optimizer: AdamW
• Learning Rate: 2e-5
• Weight Decay: 0.05

@inproceedings{ reuss2025flower, # Add citation when available }

License

This model is released under the MIT license.