Liputan6.com, Jakarta - Anthropic, a leading artificial intelligence research company, officially launched its latest AI model, Claude Opus 4.7, on Thursday, April 16, 2026.
This model represents a substantial improvement over its predecessor, Claude Opus 4.6, which was released in February 2026 and is now generally available to a variety of platforms and users.
Advertisement
The arrival of Claude Opus 4.7 is expected to revolutionize the way professionals interact with AI, especially in tasks requiring high precision and complex problem-solving.
This advanced AI model is accessible through Claude AI, the Claude API, Amazon Bedrock, Google Cloud's Vertex AI, and Microsoft Foundry.
For business and consumer users, Claude Opus 4.7 is also available in Claude for Pro, Max, Team, and Enterprise users, providing flexibility in leveraging the latest AI technologies.
Significant Improvements in Claude Opus 4.7
Claude Opus 4.7 delivers significant improvements across multiple areas, making it a more powerful and versatile AI model.
Users can now confidently delegate coding tasks that previously required close supervision to Opus 4.7.
Furthermore, Claude Opus 4.7 boasts significantly improved vision, capable of processing images with a resolution of up to 2,576 pixels on their longest edge, or approximately 3.75 megapixels.
This capacity more than triples the capabilities of previous Claude models, enabling multimodal applications that rely on fine visual details, such as reading dense screenshots or extracting data from complex diagrams.
The model also excels in long-running task handling and consistency, capable of completing complex, long-running tasks with high accuracy and consistency.
Claude Opus 4.7 is also claimed to be more sophisticated and creative in completing professional tasks, producing higher-quality interfaces, slides, and documents.
Innovative Features
Anthropic has enhanced the agent capabilities in Claude Opus 4.7.
The model can plan intentionally, use memory to learn between sessions, and encourage long-term work with minimal supervision.
An interesting new feature is the 'xhigh' effort level, which lies between the previous 'high' and 'max' settings.
Additionally, Task Budgets have been launched in public beta for API users, allowing further control over agent tasks.
Claude Opus 4.7 also uses a new tokenizer that can generate 1.0 to 1.35 times more tokens for the same input, depending on the content type.
On the security side, Anthropic reports that Opus 4.7 exhibits a low risk of misaligned behavior, with a similar risk profile to Opus 4.6, and is less likely to cause hallucinations.
Comparison with Claude Mythos and Availability
While Claude Opus 4.7 is Anthropic's most intelligent model available to the public, Anthropic states that it is not as robust as Claude Mythos Preview.
Claude Mythos Preview is considered too dangerous for wide public release due to its advanced cyber capabilities, and is currently in limited release.
Anthropic implemented cyber safeguards in Opus 4.7 that automatically detect and block requests that indicate prohibited or high-risk cybersecurity usage.
In performance benchmarks, Opus 4.7 scored highly with 64.3% on SWE-bench Pro, 87.6% on SWE-bench Verified, and 69.4% on Terminal-Bench 2.0.
The pricing for Claude Opus 4.7 remains unchanged from Claude Opus 4.6, at $5 per million input tokens and $25 per million output tokens.