k-point mesh

How finely reciprocal space is sampled. Core keeps every mesh a structure can have in one ordered ladder, with the k-distance and mesh index each one corresponds to. A model predicts one quantity on that ladder; Core finds the rung.

Quantity	What it is	Model
k_distance	Largest spacing between adjacent k-points, Å⁻¹	QRF95
k_index	Position in the ordered table of meshes	k-index forest
k_index_dense	Whether a structure needs rung 12 or above	Dense-mesh screen (draft)
k_line_density	k-points per unit reciprocal length	none
k_pra	k-points per reciprocal atom	none

The dense-mesh screen advises no setting. It serves goldilocks-data: it ranks structures nobody has calculated yet, so machine time spent extending this dataset goes where its labels are thin. Core never reads it.

Two datasets can both call a column "k-distance" and differ by a factor of 2π, and nothing in the number says which. That is why a model declares a target contract rather than a column name — and why a k-index from one ladder must not be read against another.