Recent evaluations of AI models have revealed that Anthropic's Claude 3.5 Sonnet is outperforming OpenAI's O1-pro, particularly in coding tasks and other scientific disciplines. Users have noted that the performance gap between the two models is narrowing, with some suggesting that O1-pro is superior for coding applications. Feedback indicates that while O1-pro has improved significantly compared to its previous version, O1 Preview, Claude 3.5 Sonnet remains competitive, especially for general discussions and idea generation. Users have expressed that O1-pro is the best model they have used for coding, highlighting its ability to handle complex codebases effectively. Overall, the ongoing evaluations suggest that both models have their strengths, with Claude 3.5 Sonnet excelling in certain areas and O1-pro showing marked improvements in coding capabilities.