Raycaster / evalsBack to SpreadsheetBench

SpreadsheetBench

56915

0/1Fail

SpreadsheetBench task 56915. Inspect the exact spreadsheet prompt, compare published model runs, open the agent response trace, and review grades for workbook editing.

Spreadsheet editingDual harness
ssb-56915
SpreadsheetBench
2 models
Dual (parsed + original)
PromptWorkbook objective and answer cells
ResponseOpen each trace to inspect tool use and edits
GradeCompare score and pass/fail by model

Task prompt

What the agent was asked to do

You are solving a spreadsheet benchmark task in a real workbook. Objective: Produce the correct final workbook state for the expected answer region. What matters: - Only the values in the expected answer region will be graded. - The workbook is the answer. Instructions: 1. Read the workbook and inspect the relevant data region first. 2. Infer the required result for the provided workbook instance. 3. Write the final value(s) directly into the expected answer region. 4. Do not rely on prose, formulas in your chat response, pseudocode, or VBA as the answer unless the benchmark explicitly requires those to be written into cells. 5. If the natural-language task asks for a general method, formula, or macro, convert that into the concrete result needed for this workbook instance. 6. Keep your final text response short and only summarize the workbook cells you changed. Relevant data region(s): Expected answer region(s): 'COVER'!B18:E23 Expected answer sheet(s): Task: Populate cells in the range B18:E23 corresponding to the column header in Column A and row header in Row 13 on the COVER sheet with the correct data from the DATA sheet, where the entity is selected in cell B12 of the COVER sheet. The formulas I found only work for column B and don't seem to work for columns C, D, and E, which may be due to the presence of merged cells on the DATA sheet. All the data needs to be rounded off to two decimal places. Columns C and E to be updated in decimal format with negative numbers in parentheses even when they round off to zero.

Published trajectories

Agent runs on this task

Curated dual-harness runs. Best scored run per model.

ModelHarnessScoreResultLinks
GPT-5.4showcasedual0/1Fail
Gemini 3.1 Produal0/1Fail