Claude Fable 5.2 ranked first on Artificial Analysis’s VIALS-AA benchmark, scoring 67%, ahead of Gemini 4 Carbon at 64% and GPT-6 Astra at 63%. Tool use nearly doubled the average scores of all models tested. The smaller and less expensive Claude Haiku 5.2 even surpassed Claude Fable 5.2 when it had access to tools and Fable did not. The benchmark contains 161 tasks across eight domains, using 174 images drawn from real biotechnology workflows, while per-task costs varied by as much as 99 times.
Source: Read the original article
Disclaimer: this content is for information purposes only and is not financial advice. Cryptocurrencies are highly volatile: you may lose all of your capital. Always do your own research. Legal notice
