claude-opus-4-7-release

TL;DR

Anthropic has launched Claude Opus 4.7, an AI model with enhanced software engineering and vision capabilities.
It's now accessible via API, Amazon Bedrock, Google Cloud's Vertex AI, and Microsoft Foundry.
Pricing remains the same as Opus 4.6 at $5/million input tokens and $25/million output tokens.
The model shows improved performance in coding, complex reasoning, and image understanding.

Advanced Software Engineering: Opus 4.7 excels at complex, long-running coding tasks, demonstrating rigor, consistency, and self-verification.
Enhanced Vision: The model offers substantially better image resolution and produces higher-quality professional outputs like interfaces and slides.
Cybersecurity Safeguards: Opus 4.7 includes features to detect and block high-risk cybersecurity uses, with a Cyber Verification Program for legitimate research.
Performance Gains: It achieved a 13% improvement on a 93-task coding benchmark over Opus 4.6 and solved tasks previously unsolvable by earlier models.

Consistent Rates: Claude Opus 4.7 maintains the same pricing structure as its predecessor, Opus 4.6.
Input Tokens: The cost for input tokens is $5 per million.
Output Tokens: The cost for output tokens is $25 per million.

Claude Opus 4.7 represents a significant step forward for Anthropic's AI offerings; particularly in the realms of software engineering and multimodal understanding...
Anthropic has also integrated crucial cybersecurity safeguards into Opus 4.7; While not as advanced as their most powerful models; these features are designed to prevent misuse and support responsible AI deployment...
Early testing and benchmarks highlight Opus 4.7's advancements; On a challenging 93-task coding benchmark; it demonstrated a substantial leap in performance; outperforming its predecessor and even solving tasks that had stumped previous versions...