Hierarchical Reasoning Model
Open paper
Benchmark
Model
Rank
Results
sudoku-solving-on-sudoku-extreme
HRM
#4
Accuracy: 55.0
abstract-reasoning-on-arc-agi
HRM
#17
Accuracy: 40.3
abstract-reasoning-on-arc-agi-2
HRM
#19
Accuracy: 5.0
Rank counts only results with a code link.