Anthropic's Best Model Captures Only 11% of Its Own Platform Spending

Despite Anthropic's Claude Fable 5 topping AI coding benchmarks and being rated the world's best model by prediction markets, enterprise customers are largely avoiding it in favour of cheaper alternatives. According to spending data from fintech firm Ramp, which tracks corporate card transactions across 70,000 businesses, Fable 5 accounts for just 11% of enterprise spend on Anthropic's platform. The older, lower-cost Opus 4.8 dominates with roughly 50% of spend, having surged from 15% in early June to over 50% by late July 2025. Fable 5 costs double the price of Opus 4.8 and Opus 5, but most corporate AI workloads — such as summarisation, classification, and document extraction — do not require its added capability. The trend highlights a growing gap between building the most powerful AI model and building the most commercially successful one, even as Anthropic's annualised revenue reached $65 billion in July.