How can a 7M-parameter recursive model outperform much larger models on ARC-AGI?

Understand the nested recursion, adaptive computation, and Q-head behind the 7M-parameter Tiny Recursive Model.

How a 7M parameter model beats 1T models at Sudoku, Mazes, and ARC-AGI.

TRM Architecture: Uses recursive computation with three nested loops: Innermost Loop (latent recursion): Phase A (reasoning, updating z) + Phase B (refining answer y) Middle Loop (deep recursion): 2 warm-up rounds without gradients + 1 final with gradients Outermost Loop: Up to 16 repetitions with adaptive computation time via learned Q head

The key insight: small models can match or beat massive ones through recursive depth instead of parameter count.