JAXLEY open-source software toolbox trains complex networks with up to 100,000 parameters to perform tasks like memory and vision

https://phys.org/news/2025-11-software-toolbox-enables-brain.html

"bridges biophysics and machine learning inspired techniques to accelerate/ improve brain accuracy modeling... trains directly on experimental data... models use 1,000s of equations simulating electrical signaling in neurons, calculating how changes in model parameters affect simulation outcome, model self-adjusting based on data... analyzes large datasets in parallel using AI-class GPU... faster and more accurate simulations of brain function, learning, and memory"