› Updated: 9 November 2026Claude Fable 5.2 launched on 9 November 2026 as Anthropic's strongest generally available model, and the obvious question is how it stacks up against the two models many developers actually use every day: OpenAI's GPT-6.1 Sol and Google's Gemini 4 Argon.
› There is a catch: the three companies do not publish results on the same set of benchmarks, and Anthropic's own launch table compares Fable 5.2 with GPT-6 Astra and Gemini 4 Carbon, not with Sol or Argon.
› So below I separate what each vendor says, where the numbers come from, and what you pay.
› Every benchmark figure is labelled with its source.Prices: Fable 5.2 vs GPT-6.1 Sol vs Gemini 4 ArgonModelInput / 1M tokensOutput / 1M tokensCached input / 1MSourceClaude Fable 5.2$8$40$0.20Anthropic pricing docsGPT-6.1 Sol (up to 272K input)$2$10$0.10OpenAI API pricingGPT-6.1 Sol (over 272K input)$4$15$0.20OpenAI API pricingGemini 4 Argon (until 30 Nov 2026)$2$10$0.10Google AI Studio pricingGemini 4 Argon (from 1 Dec 2026)$4$20$0.20Google AI Studio pricingPer token, Fable 5.2 is four times the price of GPT-6.1 Sol and of Gemini 4 Argon at its introductory rate, and twice Argon's price from December.