Claude Sonnet 5.5 narrows the gap with premium models at lower cost

What's this about?

Anthropic released a faster everyday-work model that approaches Opus 5.5 on several coding and office benchmarks while retaining Sonnet pricing. It is available across Claude products and the major cloud platforms.

What exactly does “30% less per task” mean?

The listed token price did not fall. Anthropic says the model often completes the same task with fewer tokens and runs more than 30% faster, so the total bill for a completed job can be lower.

What's the community saying?

1. Developers focused on its unusually large Terminal-Bench improvement and near-Opus performance in Cursor's evaluation.
2. Cost-sensitive users highlighted that it became the Claude free-tier model.
3. Skeptics noted that vendor benchmarks vary by effort setting and Anthropic itself still finds Opus better for open-ended judgment.