Sonnet 5.5 keeps token prices steady while targeting cheaper completed tasks
Anthropic's September 28 release claims faster output and fewer tokens per task. Its $2 input and $10 output rates remain the same as Sonnet 5.
What happened
Anthropic introduced Claude Sonnet 5.5 on September 28 for coding, document creation and everyday professional work. The company says output generation is more than 30% faster than Sonnet 5 and that its tests show up to 30% lower cost per completed task.
The distinction that matters
Input still costs $2 per million tokens and output $10, matching Sonnet 5. The claimed savings come from using fewer tokens to finish the work. Teams should compare the total bill for successful runs, including retries and reasoning settings, rather than treat this as a reduction in the price of each token.
Migration details
The API identifier is claude-sonnet-5-5. Anthropic says deployments that used thinking off must examine the new between_tools setting. It also states that higher-risk cybersecurity tasks can fall back to Sonnet 5 under its safeguards.
The release includes stronger benchmark results, but these remain provider-reported measurements. Anthropic itself says Opus 5.5 remains stronger for complex, open-ended work requiring sustained judgment. That distinction is more useful than assuming Sonnet has replaced every higher-end model.