Grok 4.5 has reportedly outperformed Fable 5 and Opus 4.8 in an agent AI test, achieving a score of 51.4%. The implications of this result are still being assessed, with some outlets suggesting that Grok 4.5's lower cost could be a significant factor in its appeal. The development has sparked discussion about the competitive landscape of AI models, with various other models, including those from OpenAI and China's Zhipu AI, also being mentioned as part of the broader conversation.