Research
SARBLOH: Spatial Perception Engine Boosts ARC-AGI-3 Scores 13.7x
A new open-weight agent, SARBLOH, demonstrates that spatial perception, not planning, is key to interactive reasoning on ARC-AGI-3. Its zero-parameter perception engine converts raw grids into object inventories and geometric briefings, raising rubric scores from 3/28 to 25/28 and cutting token use by 89.4%. On hidden Kaggle evaluation, it scored 16.53, 13.7x the first Milestone winner, within a nine-hour budget.
ARC-AGI-3spatial reasoninginteractive agents