Skip to content
Evaluation sources

SpreadsheetBench 2

SpreadsheetBench / Renmin University / AfterQuery / Professional work · Completes spreadsheet work including modeling, workbook repair, and visualization.

Official evaluation
Across NexGateHub AI News Observing
Evidence budgetNot scored
Upstream data timeUnverified
Last successful syncCollection has not started

What it measures and how

Fixes V2's 321 tasks and common SWE-agent conditions. Independent office products and other scaffolds are not attributed to base models. Visualization correctness is not broad aesthetic quality.

How this evidence is used

Under review: stable JSON and row dates exist, but base-model coverage under common conditions is insufficient. Scores await testing of current models.

Evaluation limits & attribution

Under review: stable JSON and row dates exist, but base-model coverage under common conditions is insufficient. Scores await testing of current models.

Data license: The official footer declares MIT; dataset and code links are public.

Results are published by SpreadsheetBench / Renmin University / AfterQuery. Original scores and the NexGateHub consensus index use different scales and cannot be added directly.