Benchmark RadarAI BENCHMARK PROFILE
FrontierFinance
Comprises 220 expert-crafted queries with 11,543 source-attributed rubrics across six finance use cases, evaluating agents on open-ended analyst-style answers via rubric-based grading.
- Released
- 2026-08-12
- Readiness
- Paper only
- Primary field
- Finance & Economics
Why it matters
Captures the full investor workflow beyond narrow data extraction, offering a harder and broader evaluation that assesses agent quality and efficiency in realistic finance research.
Motivation
AI agents are increasingly deployed for professional investment research, yet no benchmark captures the complexity of the full investor workflow.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.