All AutosheetBench tasks

Convert a range to a table

Spreadsheet manipulationMediumDeterministic scoringLatest run

Passed by 9 of 12 products on their best model. Failed: Claude in Sheets, ChatGPT + Autosheet MCP and ChatGPT in Sheets.

Prompt and starting spreadsheet

Task files

Prompt

Turn this into a proper table.

Starting spreadsheet

Sheet1
ABCD
1Order IDCustomerStatusTotal
25001AcmeShipped420
35002GlobexPending180
45003InitechDelivered260
55004UmbrellaShipped540
65005WaynePending310

Expected result

How each AI configuration did

17 of 22 passRun files
Excel and Sheets · All types · All models · All outputs

Pass criteria

All must be met, checked automatically.
  • A1:D6 hold the expected values
  • A1:D6 is a table

Passed 17

Autosheet in Excel Opus 5.5 · Medium

Passed7s · 2¢

Autosheet in Excel GPT-6 Astra · Low

Passed7s · 5¢

Autosheet in Sheets Opus 5.5 · Medium

Passed18s · 5¢

Autosheet in Excel Sonnet 5.5 · Medium

Passed4s · 5¢

Claude in Excel Opus 5.5 · High

Passed10s · 6¢

Autosheet in Excel Haiku 5.5 · High

Passed5s · 0.1¢

Claude Code + Sheets MCP Opus 5.5 · Medium

Passed42s · N/A

Autosheet in Excel GPT-6.1 Sol · Medium

Passed11s · 1¢

Claude Code + Autosheet MCP Opus 5.5 · Medium

Passed45s · N/A

Autosheet in Sheets Haiku 5.5 · High

Passed20s · 0.2¢

ChatGPT in Excel GPT-6 Astra · Medium

Passed14s · 35¢

Claude in Excel Haiku 5.5 · High

Passed6s · 0.3¢

Autosheet in Excel GPT-6 Luna · High

Passed9s · 0.3¢

Copilot in Excel Auto

Passed31s · N/A

ChatGPT + Drive MCP GPT-6 · Medium

Passed26s · N/A

ChatGPT in Excel GPT-6 Luna · High

Passed34s · 0.6¢

Gemini in Sheets Auto

Passed1m 11s · N/A

Failed 5

Claude in Sheets Opus 5.5 · Medium

Failed46s · 10¢

Why it failed: A1:D6 isn't a table.

ChatGPT + Autosheet MCP GPT-6 · Medium

Failed45s · N/A

Why it failed: A1:D6 isn't a table.

Claude in Sheets Haiku 5.5 · High

Failed33s · 2¢

Why it failed: A1:D6 isn't a table.

ChatGPT in Sheets GPT-6 Astra · Medium

Failed54s · 49¢

Why it failed: A1:D6 isn't a table.

ChatGPT in Sheets GPT-6 Luna · High

Failed1m 41s · 0.8¢

Why it failed: A1:D6 isn't a table.