An experiment found that in coding tasks an AI model with a lower token price can end up costing more to fix bugs. A Fiddler research team compared Anthropic's Claude Haiku 4.5 and Claude Sonnet 5 on ...