How I AI · Tuesday, September 22, 2026
Anthropic's Opus 55 model has demonstrated strong performance in recent benchmarks, scoring higher than both GPT-4 and GPT-4 Turbo. It also outperformed Fable 1 by a significant margin.
“Opus 55 is actually doing really well it's scoring higher than GPT 4 higher than Claude 2 and it's doing better than GPT 4 Turbo which is pretty impressive”
“and then um that it's scoring significantly higher than Fable 1 which is really interesting”