Claude Opus 5 scores 30x ahead of GPT-5.6 Sol on ARC-AGI-3, priced like its predecessor.
Anthropic's Claude Opus 5 scored 30.2% on the memorization-resistant ARC-AGI-3 benchmark, nearly four times GPT-5.6 Sol's 7.8%, while holding pricing flat at $5/$25 per million tokens. The model also posted a 21-point SWE-bench Multimodal gain over Opus 4.8 in a single generation.
claude5.ai/news/claude-opus-5-benchmark-results-analysis