Claude Opus 5 sets a new high score on ARC-AGI-3. As of July 24, 2026, Claude Opus 5 (High) is the highest-performing model on ARC-AGI-3, scoring 30.2% (30.16% table). It completed five additional Public Demo environments that no model had previously beaten. At Max reasoning effort, Opus 5 scores 97.5% on ARC-AGI-1 and 90.4% on ARC-AGI-2 Semi-Private. Due to the short testing window, ARC-AGI-3 was evaluated only at High reasoning effort.
Verified scores: Max — ARC-AGI-1 97.5%, ARC-AGI-2 90.4%, ARC-AGI-3 —; High — ARC-AGI-1 97.5%, ARC-AGI-2 88.3%, ARC-AGI-3 30.16%.