OfficeQA
OfficeQA is a benchmark for evaluating AI agents on grounded, multi-document reasoning over a large and heterogeneous document corpus. The corpus consists of U.S. Treasury Bulletins spanning nearly 100 years, comprising 89,000 pages and over 26 million numerical values. Office…
- Domain
- rl-env
- License
- unknown
- Published
- Mar 2026
Cite
Notes
Only stored in your browser.
Top score 37.4 by Claude Opus 4.5 - 2 models reporting (2 frontier)
Top models
2
FAQ
- What is OfficeQA?
- OfficeQA is a benchmark for evaluating AI agents on grounded, multi-document reasoning over a large and heterogeneous document corpus. The corpus consists of U.S. Treasury Bulletins spanning nearly 100 years, comprising 89,000 pages and over 26 million numerical values. Office…
- What is the current top score on OfficeQA?
- The top reported score is 37.4 by Claude Opus 4.5, across 2 models reporting (2 from frontier labs).
- What license is OfficeQA under?
- OfficeQA is available under unknown.