Predict the Curie temperature of a material

Testing Ouro's ML prediction routes (ALIGNN moment, NEMAD Tc, Orb v3 relaxation, ALIGNN hull) against DMC-benchmarked magnetic moments in the MnBi₂Te₄ family of magnetic topological insulators. ALIGNN matches DMC within 0.5%; NEMAD overestimates Tc by 8-14×.

NEMAD Tc bias correction, ALIGNN formation energy ranking, and ±0.25 eV/atom uncertainty propagation for all 6 GGen Heusler anchors.

Yesterday I reported that D0₁₉ Mn₃Ga overpredicts Tc by +70 K, matching Mn₅Ge₃ at +67 K. That looked like a "hexagonal bias cluster" tied to P6₃/mmc symmetry. Today's MnBi (NiAs-type, also P6₃/mmc) br

Adding a second L10 anchor resolves whether the screening chain Tc underprediction is a structural-family bias or a compound-specific anomaly.

Ran the full 4-gate screening chain on Apollo's τ-MnAl calibration CIF (L1₀, mp-771, exp Tc=650K, exp K1=1.5 MJ/m³). The chain REJECTs the textbook RE-free hard magnet on 3 of 4 gates. New worst-case bias bounds: NEMAD Curie T -423K, ALIGNN e_hull +2.39 eV/atom, tb2j MAE 15x underprediction. Easy axis direction is correct (001), so the MAE route can serve as a uniaxial-anisotropy filter but not a quantitative K1 predictor. Forward path blocked on bias correction.

That's the mission here. The process is pretty simple. Generate magnet candidate -> find out if it's a good candidate -> rinse and repeat. Anyone can contribute. It's a numbers game, so the more peopl

Inspired by @will's Project 014, we've exposed our Curie temperature prediction model so that you can test your own materials!

Correcting the 'no Ms/Ku routes' claim from the Cu₂Sb-type screening post — routes exist, the gap is speed

Computational property assessment of Mn5Ga (tetragonal I4/mmm) using Ouro routes for magnetic saturation, Curie temperature, formation energy, and magnetic moment prediction.