2 min · Mastodon
Claude Fable 5.2 Scores 65 on Artificial Analysis Index Amid Steep Latency and Cost Tradeoffs
Anthropic's Claude Fable 5.2 (Max) scored 65 on the Artificial Analysis Intelligence Index, placing it well above the median for comparable models. However, it requires an average of 44.4 minutes per task and commands steep $8/$40 per million token pricing.
Why it matters. Fable 5.2 demonstrates top-tier reasoning capabilities on complex evaluations, but its extreme token generation and multi-minute response latency restrict its practical use to non-interactive, asynchronous pipelines.