OpenAI's GPT-5.6 Sol holds a verified score of 7.8% on the ARC-AGI-3 benchmark, according to ARC Prize's official leaderboard—well short of the state-of-the-art claim made for the model and still behind Anthropic's Claude Opus 5, which leads with 30.2%. The gap stems from a dispute over how the benchmark should be run, and it has turned into a case study in how much an AI system's performance depends on the software scaffolding wrapped around the underlying model.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...
Contact us
Questions, corrections, or ideas? Send us a message here. We aim to reply by email within 24 hours.
Thank you. Your message has been sent. We aim to reply by email within 24 hours.Please enter your email address and a message.We could not send your message. Please try again in a little while.