Anthropic introduced Sonnet 5.5 as a speedier, more affordable follow up to its previous Sonnet, available starting Monday across all platforms. The company priced it at $2 per million input tokens plus $10 per million output tokens, noting this is half the price of Opus 5.5. It also said Sonnet 5.5 typically uses fewer tokens per task than the earlier version, which lowers overall operating costs.
Performance wise, Anthropic says Sonnet 5.5 improves on coding, handles scoped tasks more reliably, and produces higher-quality documents, slide decks, and spreadsheets compared with the prior version.
This release lands under a week after Anthropic rolled out Opus 5.5, the pricier model in the family, and the company said Haiku 5.5, the most affordable tier, is coming soon. On who should choose Sonnet, CNBC heard from Theo Chu, who serves as a research product manager with Anthropic: "Sonnet is really for the cost-conscious customer where they might not need as much intelligence." He added, "It might be routine tasks that just need execution, but don't need that judgment that Opus can bring."
Sonnet 5.5 is live on Amazon Web Services, Google Cloud, and Microsoft Azure.
When innovations reshape industries, a steady plan helps protect and grow your savings. Join Briefs Finance CEO Jaspreet Singh on September 29th for a FREE live investor workshop, How to Profit From A Dollar That's Losing its Value, where he shows how we're spotting investment opportunities as the dollar falls. Save your spot.
Anthropic noted that Sonnet 5.5 isn't stretching the limits of its models' abilities, so most alignment evaluation concentrated on a "targeted set of risks that apply to models of any capability level." In AI, alignment typically refers to a model behaving in ways that reflect human interests and values.
Cyber protections are a bigger deal this time. Anthropic said Sonnet 5.5 is a "large improvement" in cybersecurity compared with the prior Sonnet and marks the first Sonnet release that ships with similar fallback mechanisms and cyber safeguards Anthropic developed for its most capable models. "Focusing on alignment and safety has been a key part of our mission from the very beginning," Chu said.
All of this comes as scrutiny of AI safety has intensified, with more researchers warning that advanced AI could cause catastrophic harm. Earlier this month, Amodei called on AI companies to reduce the pace at which they improve their most advanced models.
For teams and solo builders watching usage costs, Sonnet 5.5 targets routine, execution heavy work at a lower price, with added efficiency from using fewer tokens per task. If you are buying compute by the token, that combo can bring the bill down while still getting stronger coding and doc creation than the last Sonnet.
Keeping a clear financial strategy makes adapting to new tools less risky. Our CEO Jaspreet Singh is hosting a FREE live investor workshop, How to Profit From A Dollar That's Losing its Value, on September 29th. Sign up free to join him live.
