GDP.xlsx: Benchmarking Frontier Models on Professional Workbooks
Factual evidence
What the source reports
Researchers introduced GDP_xlsx, a benchmark of 70 complex spreadsheet tasks spanning 13 domains to evaluate AI agent performance.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- arXiv cs.CL — Computation and Language · RESEARCH
- Published by source
- 16 November 2026
- Collected by OneBench
- 17 Nov 2026, 02:02 UK
- Original headline
- GDP.xlsx: Benchmarking Frontier Models on Professional Workbooks ↗
Stored source excerpt
arXiv:2611.17766v1 Announce Type: new Abstract: We introduce GDP_xlsx, a benchmark of 70 realistic workplace requests written by professionals in 13 domains, each paired with an…
Short excerpt from the collected text, not the full source. Use the source link to read it in context.
The factual summary is a OneBench synthesis, not a quotation or independent verification. Collection time is not publication time. Open the source for its full context; related reporting can share the same underlying announcement.
OneBench interpretation
Institutional assessment
So what
Benchmarking AI agents on complex, multi-tab spreadsheets provides empirical validation criteria for automating analytical financial workflows.
Do what
Review model evaluation methodologies with quantitative research and data science teams testing spreadsheet-automation agents.