Anthropic launches Claude Sonnet 5.5, a faster, cheaper AI model

Anthropic on Monday released Claude Sonnet 5.5, the second model in its Claude 5.5 family and the replacement for Sonnet 5, which shipped in June . The company calls it a clear upgrade that runs more than 30% faster and costs up to 30% less for most work, alongside higher benchmark scores . Anthropic says the model is strongest at scoped everyday tasks such as fixing bugs and producing polished documents, slides and spreadsheets .

The cost claim comes from efficiency rather than a price cut. Sonnet 5.5 keeps Sonnet 5's list pricing of $2 per million input tokens, $10 per million output tokens and $0.20 per million cache reads, but Anthropic says it typically needs far fewer tokens to finish the same work . That list price is half of Opus 5.5's, and research product manager Theo Chu told CNBC that Sonnet "is really for the cost-conscious customer" whose routine tasks need execution more than Opus-level judgment . The model is available across platforms at launch, including Amazon's .

The release lands less than a week after Opus 5.5 and is Anthropic's second launch since CEO Dario Amodei urged AI companies earlier this month to slow how quickly they improve their most advanced models . Anthropic says Sonnet 5.5 does not advance its capability frontier, but its cybersecurity skills are a "large improvement" over its predecessor, so it is the first Sonnet model to ship with the cyber safeguards and fallbacks built for the company's most capable models . Haiku 5.5, the cheapest tier, is described as coming soon .

What to watch: whether the token-efficiency gains show up in enterprise usage and pricing pressure on rivals' mid-tier models, how quickly Haiku 5.5 follows, and whether the steady release cadence under a stated slowdown posture shapes how investors read Anthropic's widely reported IPO plans.