Claude Fable 5: $6K Benchmark Highlights High AI Costs

Anthropic's new AI model, Claude Fable 5, costs over $6,000 just to run its benchmark tests. This flagship reasoning model achieved the top spot on the Artificial Analysis Intelligence Index. Here's the thing: The benchmark evaluation consumed 87 million output tokens. Claude Fable 5 is priced at $10 per million input tokens and $50 per million output tokens. This is why the final bill reached $6,227.74. What's interesting is that reasoning models produce many more output tokens than standard chat models. This is because they work through complex problems step by step. On the SWE-Bench Pro, Fable 5 scored 80.3% accuracy, a significant improvement over its predecessor and competitors. The model also handles both image and text inputs and has a 1 million token context window. The bottom line is that advanced AI capabilities come with a high price tag.