Anthropic Launches Claude Sonnet 5.5 As AI Coding Race Accelerates

Anthropic launched Claude Sonnet 5.5 on Monday, just days after introducing its higher-end Opus 5.5, as the AI company pushes to make increasingly capable models cheaper and faster for software development and other workplace tasks.

The release adds two new models to Anthropic’s Claude 5.5 lineup within a week. Opus 5.5, unveiled Sept. 22, was aimed at more complex work and coding, while Sonnet 5.5 is designed for more routine tasks that still require substantial reasoning.

Anthropic said Sonnet 5.5 generates output more than 30% faster than Sonnet 5 and can cost up to 30% less per task. But the company isn’t cutting its API prices: input remains $2 per million tokens and output $10 per million tokens. The claimed savings instead come from the model using fewer tokens and tool calls to complete tasks.

That efficiency push comes as AI companies increasingly compete not just on benchmark scores, but on the cost of running coding agents through long, multi-step tasks. Anthropic’s Opus 5.5 launch last week similarly emphasized lower operating costs, with the company saying the model was 40% cheaper to run than its predecessor.

Sonnet 5.5 also showed gains outside software development. Anthropic reported a score of 1,811 on AA-Briefcase v1.1 and 61.6% on Chartography without tools. The company said the model is also the first Sonnet version to beat "Pokémon Red" using screenshots alone.

Anthropic is adding stronger cybersecurity safeguards alongside the performance gains. The company said Sonnet 5.5 is the first Sonnet model to launch with cyber protections and fallback measures similar to those used for its more capable systems.

Early testers cited by Anthropic included Daniel Vogel, chief operating officer at Epic Games, who said the model handled tens of thousands of lines of gameplay-system code as well as system-design and data-flow reviews.

Anthropic said Claude Haiku 5.5 will follow in the coming weeks, giving the company a three-model 5.5 lineup spanning higher-end reasoning, general-purpose work and high-volume applications.

Photo: Shutterstock