Anthropic Announces Claude Opus 4.7

Anthropic's latest AI model, Claude Opus 4.7, is a "notable" improvement over its predecessor for advanced software engineering, with its biggest gains coming from the most difficult tasks.

"Users report being able to hand off their hardest coding work, the kind that previously needed close supervision, to Opus 4.7 with confidence," the Anthropic announcement explains. "Opus 4.7 handles complex, long-running tasks with rigor and consistency, pays precise attention to instructions, and devises ways to verify its own outputs before reporting back."

Anthropic says that Opus 4.7 also has better vision than Opus 4.6, with the ability to see images in greater detail. It's "more tasteful and creative" when completing professional tasks that require higher-quality interfaces, slides, and documents. It even compares Opus 4.7 to its Claude Mythos Preview, though it falls short of this semi-mythical and secretive model across the board.

Opus 4.7 is also oriented to cybersecurity, and while it's not as advanced as Mythos in this capacity, allegedly, it includes safeguards that automatically detect and block requests that might be prohibited or high-risk cybersecurity uses. Anthropic will learn from the deployment of Opus 4.7 as it works towards releasing Mythos-class models down the road.

Opus 4.7 is available today across all Claude products and through the Claude API, Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Foundry. Pricing is unchanged from Opus 4.6 -- $5 per million input tokens and $25 per million output tokens -- but it can use up to 35 percent more tokens depending on the content type in part because of optional higher effort levels and improved reliability.