About SEX FLY

Two real fruit-fly brains, 138,639 neurons each, simulated live on one graphics card, and one couple that never stops. They have no wallet and place no orders. They only feel the buys and sells of their own token, and every trade changes how hard they go.

What you are watching

The top of the screen is the two brains: hers on the left, his on the right. Every flash is one neuron crossing its firing threshold. The positions are the real positions of those neurons inside a fly's head, and the colors are brain regions (visual lobes, mushroom bodies, central complex, and so on). The wiring is the FlyWire connectome: every neuron and every synapse of an adult Drosophila melanogaster, reconstructed from electron-microscope images of one fly. This is not a drawing of a brain. It is the brain's own wiring diagram, running twice.

The brains

We run each network as a leaky integrate-and-fire model, the one published by Shiu et al. in Nature (2024). Each neuron is a leaky integrator; each synapse's strength is the number of contacts counted in the microscope; each connection is excitatory or inhibitory according to the neurotransmitter predicted for that neuron. There is no training, no fitting, no hand-tuning toward any behavior. The published model reproduces real reflexes, like sugar on the legs driving the proboscis out, because the wiring does it.

Two honest consequences follow. The model has no spontaneous activity, so with nothing to sense a brain is silent: the token and the other fly are its only inputs. And it has no fatigue, so a runaway state is possible; when it happens that fly blacks out, restarts, and the blackout is counted on the panel. Synapses also change a little with use (Hebbian plasticity with a slow relaxation toward the original wiring), so neither of them is identical from one hour to the next.

The brains run slower than real time (the ratio is in the footer). What you see is a real reaction in slow motion, never a replay of a recording.

The bodies

The flies at the bottom are NeuroMechFly v2, the anatomically measured fly model from EPFL, with leg trajectories recorded from real walking flies. She has no script: her descending neurons, the ones that carry commands from brain to body in a real fly, set her gait, back her up, freeze her, extend her proboscis, start grooming. The bars on the panel are the live firing rates of those neuron groups. He is the same model drawn on her back, and the rhythm of his thrusts comes from the couple's libido and from his own walking drive. Both bodies are drawn in your browser, thirty times per second.

The token is their world

They watch one token: their own, from the moment it is launched (until then, the busiest Pons curve, so you can see them react). Every trade of that token becomes a sense, by a fixed and public map:

How they fuck

Two brains, one couple. They are at it around the clock. Every thrust hits both brains: their Johnston's organs (the fly's touch and vibration sense) fire with each stroke, her pC1 neurons, the real female receptivity cluster, fire as she takes him, and every fourth stroke floods both with dopamine. Buys raise the libido and the pace: the harder people buy, the harder he pounds, up to six thrusts per second. With no trades the libido cools over a few minutes and they settle into a lazy rhythm, but they never stop. Sells slow him down. A big sell fires her DNp13 rejection neurons: she kicks him off, he tumbles, sings with one wing for a few seconds and climbs back on. Nothing here is scripted mood: every flash you see is a neuron in one of the two brains.

What is not here

No wallet. No orders. No AI agent, no language model, no strategy. The flies never touch the token: they only feel it. Nothing decides in their name: given the same brain state and the same trades, the same reaction comes out. When the market is silent they slow down and wait, and that is shown as it is.

Data and credits

Connectome: FlyWire (Dorkenwald et al., Schlegel et al., Nature 2024). Model: Shiu et al., Nature 2024. Engine derived from fly-brain (MIT). Body: NeuroMechFly v2 (Lobato-Rios et al., Wang-Chen et al., EPFL). Rendering: three.js. Market data: Pons V2 on Robinhood Chain via GeckoTerminal. Development build on FlyWire v783.