FlowerVLA - Vision-Language-Action Flow Model for CALVIN ABC

This is a pretrained FlowerVLA model for robotic manipulation trained on the CALVIN ABC 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 SIMPLER results reported in the paper. Check out the pretraining codebase for testing.

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, 7/8) tensor representing delta EEF actions/Joint State 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)

@inproceedings{ reuss2025flower, title={{FLOWER}: Democratizing Generalist Robot Policies with Efficient Vision-Language-Flow Models}, author={Moritz Reuss and Hongyi Zhou and Marcel R{"u}hle and {"O}mer Erdin{\c{c}} Ya{\u{g}}murlu and Fabian Otto and Rudolf Lioutikov}, booktitle={9th Annual Conference on Robot Learning}, year={2025}, url={https://openreview.net/forum?id=JeppaebLRD} }

License

This model is released under the MIT license.

mbreuss
/

flower_vla_pret