Evaluates AI agents on multi-step financial analysis tasks: earnings analysis, portfolio optimization, and structured report generation.
This definition has not completed editorial and provenance review. Verify its metric, version, source, and evaluation configuration upstream before using it for a model-selection decision.
Tests domain-specific agentic capability in finance. Multi-step workflows require planning and numerical reasoning.
Introduced by Anthropic for Opus 4.7 launch — limited third-party validation. Version 1.1 makes cross-version comparisons difficult.
Higher is better. Task completion accuracy across financial analysis workflows.
Max score: 100
Introduced by Anthropic for Opus 4.7 launch — limited third-party validation.
| # | Model | Score |
|---|---|---|
| 1 | 64.4 | |
| 2 | 61.5 | |
| 3 | 60.1 | |
| 4 | 59.7 |