• Highest capability in the family: described as “our most powerful model yet” by Anthropic.
• Exceptional at long-running tasks requiring thousands of steps and sustained focus (e.g., continuous codebase work for hours).
• Excellent performance on benchmarks: e.g., SWE-bench 72.5 % and Terminal-bench 43.2 %.
• Designed for complex agentic workflows, deep reasoning, tool use, and large context windows.
• Placed under a higher safety classification (ASL-3) due to its frontier capability and risk profile.

1. Strong general-purpose reasoning

• Great for coding, analysis, creation, and multi-step tasks.

2. Stable commercial-grade model

• Predictable output quality and long-term stability.

3. Supports batch operations

• Batch inference is 50% cheaper.

4. Good for production agents

• Reliable instruction following and structured output.