Skip to content
← Playground

Fly Brain

Paint a brain. Watch a fly. Teach it what you like.

Teach the fly

Watch a behavior, then reward or discourage it. Your feedback teaches which neurons to stimulate next. The connectome stays unchanged.

Ready for its first behaviorNo trial yet

Start a trial to teach a movement preference.

Before / After uses the same random seed. Replays never train the policy.

Learned preferences

Chance of each stimulus, including 10% exploration.

0 rated trials

    01

    Paint the brain

    Paint neurons. Release to stimulate.

    Ready to load
    L
    RestingStimulatedFiring
    R
    Actions
    02

    Watch the response

    At rest
    Drosophila melanogaster1 mm
    Walking drive0 Hz
    Turning0°/s
    Escape drive0 Hz
    Ready to set up166,700 neurons
    WebGPU · auto

    Simulation runs locally.

    Fly Brain

    Load the neural weights and prepare the scene to start painting.

    166,700 neurons · 76 MB neural weights

    Runs locally in your browser. Downloaded weights are cached for next time. Back to Playground.

    Experiment settings

    Model & sources

    166,700 neurons from MaleCNS v1.0. Dataset IDs identify individual cells; they are not consecutive neuron numbers.

    Recorded connectivity and adapted spiking dynamics drive crafted walking, turning, and flight animations. The movements are illustrative, not validated predictions of fly behavior. The scanned body is a female NeuroMechFly specimen.

    Turn actions stimulate LC9 and DNa02 on the selected side. Directly stimulated neurons can still fire when recurrent transmission is disabled.

    MaleCNS data ↗Neural model ↗@huggingface/kernels ↗Model parameters ↗
    Adapted from Xenova’s Neural Canvas · Full data provenance · Licenses & credits