SpreadsheetBench
37462
SpreadsheetBench task 37462. Inspect the exact spreadsheet prompt, compare published model runs, open the agent response trace, and review grades for workbook editing.
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): F4:F244 Expected answer sheet(s): Task: In my Excel file, I've inputted country names and revenues in local currency, which are then automatically converted into USD in the order they appear in the report. I'm attempting to list all the countries and use a formula to sum the products and search to return the corresponding number for a given country. However, while my formula works when I want it to return the local currency amount, it fails when I want it to return the USD amount, which is my actual goal. In the spreadsheet, I highlighted the failing formula in red. I'm unclear why the formula for the USD column isn't working as expected. Apply a formula to all Revenue cells that will retrieve the correct USD values, rather than INR values; leave any zero or null values as 0. After applying the formula, apply a cell infill color of #FFC7CE and text color of #9C0006 to all cells with the revised formula.
Published trajectories
Agent runs on this task
Curated dual-harness runs. Best scored run per model.
| Model | Harness | Score | Result | Links |
|---|---|---|---|---|
| GPT-5.4showcase | dual | 0/1 | Fail | Share pagePublic trace |