Anthropic releases Sonnet 4.6
The new midsize model scores 60.4% on ARC-AGI-2 and now handles entire codebases in one request.
Deep Dive
Anthropic released Sonnet 4.6, its updated midsize AI model. Key specs include a 1-million-token context window (double its predecessor) and top scores on coding (SWE-Bench) and computer use (OS World) benchmarks. It becomes the default for free and Pro users. Professionals can now feed it entire codebases, lengthy contracts, or dozens of research papers in a single prompt for analysis and generation.
Why It Matters
Enables developers and researchers to process and reason over massive documents, like full software projects, without chunking.