Anthropic Releases Claude Opus 4.7 with Advanced Coding and Cyber Safeguards
Anthropic launches Claude Opus 4.7, offering major improvements in complex software engineering and vision capabilities. The new model includes automated safeguards to block high-risk cybersecurity requests as part of the Project Glasswing initiative.
Anthropic releases Claude Opus 4.7 as a generally available upgrade that significantly outperforms its predecessor in advanced software engineering. The model handles difficult, long-running coding tasks with consistency and requires less close supervision from users. It also features improved high-resolution vision capabilities and produces more creative, professional-quality interfaces, slides, and documents.
As the first model released under the Project Glasswing initiative, Opus 4.7 introduces new automated safeguards that detect and block prohibited or high-risk cybersecurity requests. While it is less broadly capable than the powerful Claude Mythos Preview, Anthropic uses Opus 4.7 to test these cyber safeguards in a real-world environment. Security professionals who need the model for legitimate vulnerability research or red-teaming can apply to the new Cyber Verification Program.
Claude Opus 4.7 is available immediately across all Claude products, the Claude API, Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Foundry. Pricing stays consistent with the previous generation at $5 per million input tokens and $25 per million output tokens. Developers can integrate the new model into their workflows starting today by using the claude-opus-4-7 identifier.