Raycaster / evalsBack to SpreadsheetBench

SpreadsheetBench

433-47

1/1Pass

SpreadsheetBench task 433-47. Inspect the exact spreadsheet prompt, compare published model runs, open the agent response trace, and review grades for workbook editing.

Spreadsheet editingDual harness
ssb-433-47
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): A1:I21 Expected answer region(s): A1:I21 Expected answer sheet(s): Sheet1 Task: How can I add to or subtract from cells across five unit columns (from unit 1 to unit 5 for addition, and from unit 5 to unit 1 for subtraction) in order to balance positive or negative residuals in another column, ensuring that the operation adheres to specified minimum and maximum quantity constraints for each unit column, and without changing unit column cells that are zero? Be careful to keep the original location of the data unchanged. And also change Column I values to 0 at the end.

Published trajectories

Agent runs on this task

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

ModelHarnessScoreResultLinks
GPT-5.4showcasedual1/1Pass
Gemini 3.1 Produal1/1Pass