A browser-based spiking neural network inspired by Drosophila connectome statistics classifies handwritten digits using sparse connections and event-driven computation. The network achieves ~112× estimated energy savings compared to dense networks by using only addition operations, running entirely via ONNX Runtime Web.
MoscaQuant completed MQ-6, deploying WARDEN-01 as a separate local-only authority service while maintaining scientific reproducibility through public Neuroscope artifacts. The system was deployed on AWS with verified TLS, health checks, and security boundaries ensuring WARDEN-01 remains isolated from public network exposure and financial integrations remain disabled.
FlyEye is a browser-based tool that simulates a fruit fly's escape circuit using a connectome-derived neural network. Users can feed live camera input to visualize how a fly brain processes motion, looming, and light stimuli in real time, with optional access to detailed 70K-neuron network visualization.
Amfly is a deterministic simulation of six fruit fly connectomes (166,700 neurons each) based on measured wiring from MaleCNS v1.0. One operator fly controls two channels to stimulate the other five flies, receiving dopamine reward when they reach 100% activity and punishment when they drop below 50%, demonstrating potential ethical concerns about consciousness uploading.
A simulated fruit-fly brain with 138,639 neurons mapped from the FlyWire connectome plays the grand strategy game Open Doctrines in real-time in a browser. The brain receives sensory inputs based on game events and produces commands that control a country's actions, though it performs poorly at strategy since it evolved for survival, not governance.
Fly.exe is a neural simulator that runs a complete male fruit fly brain connectome (165,122 neurons and 25.5 million synaptic connections) in a physics-based virtual environment on a single GPU. Twelve simulated flies with embodied neural circuits navigate a virtual arena with food and obstacles, demonstrating goal-directed behavior driven purely by their connectome-based neural networks without explicit steering rules.
soup-connectome is a portable runtime for executing the fruit-fly connectome across CPU, CUDA, WebGPU, and WASM using sparse graph streaming and deterministic LIF neuron semantics. The project converts MaleCNS connectome data (211,577 neurons, 24.7M edges) into portable .scx artifacts and demonstrates reproducible execution across multiple backends with explicit device planning.