Anthropic has unveiled Claude Sonnet 4.6, a new iteration that brings the high-tier performance of last year’s Opus 4.5 to a much more accessible price bracket. While maintaining the standard Sonnet pricing of $3 per million input tokens, the model demonstrates a significant leap in capability, effectively bridging the gap between mid-range efficiency and flagship intelligence.

The update introduces several technical refinements, including a "reliable knowledge cutoff" extended to August 2025, ensuring the model remains current with recent global events. While the default context window—the total amount of information the model can process in a single interaction—remains at 200,000 tokens, Anthropic is testing a beta feature that allows this to expand to 1 million tokens for more data-heavy tasks.

Developers are already integrating the new models into existing workflows using specialized tools. Simon Willison (a prominent technologist and developer) updated his toolset using Claude Code, an AI coding tool that handled the complex migration logic. This includes managing "adaptive thinking," a feature where the model adjusts its internal reasoning depth based on prompt complexity, and shifts in how the model handles pre-written text prefixes.

In creative tests, Sonnet 4.6 showed a quirky personality, consistently adding top hats to SVG illustrations of pelicans. While the flagship Opus 4.6 still leads in anatomical precision for complex imagery, Sonnet 4.6 offers a compelling balance of speed, cost-effectiveness, and personality for everyday generative tasks.

Anthropic has unveiled Claude Sonnet 4.6, a new iteration that brings the high-tier performance of last year’s Opus 4.5 to a much more accessible price bracket. While maintaining the standard Sonnet pricing of $3 per million input tokens, the model demonstrates a significant leap in capability, effectively bridging the gap between mid-range efficiency and flagship intelligence.

The update introduces several technical refinements, including a "reliable knowledge cutoff" extended to August 2025, ensuring the model remains current with recent global events. While the default context window—the total amount of information the model can process in a single interaction—remains at 200,000 tokens, Anthropic is testing a beta feature that allows this to expand to 1 million tokens for more data-heavy tasks.

Developers are already integrating the new models into existing workflows using specialized tools. Simon Willison (a prominent technologist and developer) updated his toolset using Claude Code, an AI coding tool that handled the complex migration logic. This includes managing "adaptive thinking," a feature where the model adjusts its internal reasoning depth based on prompt complexity, and shifts in how the model handles pre-written text prefixes.

In creative tests, Sonnet 4.6 showed a quirky personality, consistently adding top hats to SVG illustrations of pelicans. While the flagship Opus 4.6 still leads in anatomical precision for complex imagery, Sonnet 4.6 offers a compelling balance of speed, cost-effectiveness, and personality for everyday generative tasks.