Astronomers using the James Webb Space Telescope identified supernova SN 2023-aeaf at redshift 3.195, whose light traveled about 11.7 billion years, making it one of the most distant confirmed exploding stars. ^[1] Isaac Malsky and co-authors present a machine learning local-box chemical kinetics solver for exoplanet atmospheres that uses a residual flow-map architecture. ^[2] The surrogate model is several orders of magnitude faster than a classical solver, achieving microsecond-scale inference while retaining percent-level accuracy. ^[3] Based on comparisons of its light curve and colour with simulated supernova populations, the team led by Valeria Aparicio classified SN 2023-aeaf as a Type II supernova with a probability of 97.2%. ^[4] The study was published in The Astrophysical Journal on 2026-08-13, with lead author con Valeria Aparicio from the Institute for Astronomy at the University of Hawaiʻi. ^[5] The supernova's host galaxy is a young, low-mass star-forming dwarf galaxy with relatively few heavy elements, which the team said is consistent with the metal-poor environment of a cosmic star-formed star galaxy at a similar redshift. ^[6] The model outperforms several commonly used machine learning architectures and performs robustly under the extreme stiffness characteristic of atmospheric chemistry. ^[7] The surrogate model covers a parameter space spanning T=300-3000 K, P=10^-6 to 10^4 bar, Δt=10^-3 to 10^8 s, and compositions from 10^-2 to 10^3 times solar in both C/O ratio and metallicity. ^[8]