According to The Decoder, Anthropic is using an unpublished internal model called "Model 2." Based on the available description, it outperforms any publicly available version of Claude.
The significance of this news lies in the potential gap between Anthropic's internal and public capabilities. If the description is accurate, the company is testing or deploying a system not yet offered to users.
Details regarding the model's specifications, release timeline, and evaluation methods are absent from the provided materials. The source is not a statement from Anthropic, and available confirmation is limited to metadata and a brief synopsis of The Decoder's publication.
editorial commentary
Why it matters
A likely consequence is increased attention to the difference between Anthropic's internal developments and its public products. The next observable signal will be an official statement from the company or the emergence of verifiable information regarding the model's characteristics and availability. Significant uncertainty remains due to the lack of primary confirmation and details.