AIThe Decoderabout 2 hours ago
OpenAI GPT-5.6 Sol beats Anthropic Opus 5 on ARC-AGI-3
1 min read
OpenAI's GPT-5.6 Sol scored 38.3 percent on the ARC-AGI-3 benchmark using its Responses API with Retained Reasoning and Compaction, surpassing Anthropic's Claude Opus 5 at 30.2 percent. In the official test harness without those aids, GPT-5.6 Sol scored only 7.8 percent.
Level
Hype check
Tap to vote and see what everyone thinks.
