Anthropic has released a new version of its mid-size Sonet model, keeping up with the company’s four-month update cycle. In a post announcing the new model, Anthropic emphasized improvements in coding, instruction-following, and computer usage.
Sonnet 4.6 will be the default model for free and pro plan users.
The beta release of Sonnet 4.6 will include a reference window of 1 million tokens, twice the size of the largest window previously available for Sonnet. Anthropic described the new reference window as “large enough to house entire codebases, long contracts or dozens of research papers in a single request.”
The release comes exactly two weeks later Opus 4.6 launchedAn updated Haiku model is likely to follow in the coming weeks.
The launch comes with a new set of record benchmark scores, including OS World for computer usability and SWE-Bench for software engineering. But perhaps most impressive is its 60.4% score on the ARC-AGI-2, which aims to measure skills specific to human intelligence. The score puts the Sonnet 4.6 above most comparable models, although it still lags behind models like the Opus 4.6, Gemini 3 Deep Think, and a refined version of GPT 5.2.
