Raycaster / evalsTask overviewSpreadsheetBenchCategory

SpreadsheetBench · GPT-5.4 · dual

50250

1/1Pass

GPT-5.4 on SpreadsheetBench: 50250 (dual harness). Browse score, rubric, and public trace.

1/1 · Pass
SpreadsheetBench
AI Agents for Spreadsheet Automation and Workbook Editing

Prompt excerpt

Task context

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): H2 Expected answer sheet(s): Task: Use the value in cell G2 ('sela') as a reference to sum the numbers of the values corresponding to G2 in columns A or B in C-E, and include them only if the cell's value is a number. Place it in H2.

Response trace

Agent response, tools, files, and edits

Open full trace

On mobile, open the full trace for the best view of the agent response, tool calls, file reads, edits, and grading context.

Open mobile trace