How are the baryons and dark matter distributed within massive galaxy clusters? We present an ongoing project to reconstruct the mass distribution of galaxy clusters using cosmological simulations and machine learning. We introduce HYPER, a hydro-particle code for efficient and rapid simulations of gas and dark matter. The novel approach to use subgrid models for the thermodynamics of the ICM and IGM results in three orders-of-magnitude speedup compared to a standard hydro code. In addition to the particle data, HYPER also produces lightcone catalogs of dark matter halos and full-sky tomographic maps of the lensing convergence, SZ effect, and X-ray emission. We discuss how generative diffusion models can be trained on multi-wavelength images of galaxy clusters to predict the projected gas and dark matter density fields. We some preliminary results that demonstrate the potential to map the unknown in galaxy clusters. How are the baryons and dark matter distributed within massive galaxy clusters? We present an ongoing project to reconstruct the mass distribution of galaxy clusters using cosmological simulations and machine learning. We introduce HYPER, a hydro-particle code for efficient and rapid simulations of gas and dark matter. The novel approach to use subgrid models for the thermodynamics of the ICM and IGM results in three orders of magnitude speedup compared to a standard hydro code. In addition to the particle data, HYPER also produces lightcone catalogs of dark matter halos and full-sky tomographic maps of the lensing convergence, SZ effect, and X-ray emission. We discuss how generative diffusion models can be trained on multi-wavelength images of galaxy clusters to predict the projected gas and dark matter density fields. We some preliminary results that demonstrate the potential to map the unknown in galaxy clusters.