AI Spreadsheet Benchmark

Spreadsheet-assistant benchmark with realistic spreadsheet prompts, dynamic-output checks, and latency measurements.

5rows
pass_at_1primary metric
2026-05-27sampled

Metadata

Metrics

Pass@1, Pass@3, Dynamic Output Rate, Mean Time (lower is better)

Latest Results

Rows parsed from the public Hugging Face dataset card baseline-results table. The benchmark covers 53 spreadsheet prompts.

Rank Subject Pass@1 Model Match Provenance Sampled
1 Rows AI Analyst 89% Imported 2026-05-27
2 Microsoft Excel Copilot 53% Imported 2026-05-27
3 Google Sheets + Gemini 57% Imported 2026-05-27
4 Shortcut 83% Imported 2026-05-27
5 Julius 75% Imported 2026-05-27