Model chooser  /  Analyse spreadsheets and numbers  /  One-off

To analyse one spreadsheet, upload it to GPT-6 Astra in ChatGPT.

A one-off analysis is a conversation: upload the file, ask, look at the chart, ask again. Astra is OpenAI's most capable model, it runs code on your file inside ChatGPT, and it ties for the top Intelligence Index score.

How to use it ↓ Try another task

Why GPT-6 Astra

The reasons it wins for this job.

  • Astra is OpenAI's model for "the hardest end-to-end work", with documents and reasoning among its stated strengths. Released 3 September 2026.
  • It ties Claude Fable 5.1 for first place on the Artificial Analysis Intelligence Index, which weights maths and reasoning heavily.
  • ChatGPT runs Python against the uploaded file, so the numbers come from computation rather than from the model reading figures and guessing.
  • It draws the chart, and you can download the cleaned file, in the same thread.

How to use it

4 steps to a first result.

  1. 1Upload the raw exportGive it the original file, and say what each column means if the headers are cryptic.
  2. 2Ask it to check the data firstMissing values, duplicates, date formats. Most wrong analyses are wrong data.
  3. 3Ask the business question"Which channel brought customers who stayed past 90 days" beats "analyse this".
  4. 4Ask for the codeHave it show the calculation. Check one number by hand.

Prompt for ChatGPT

Start from this.

Edit the parts in capitals, then run it.

The attached file is WHAT IT IS, exported from SYSTEM on DATE. Columns: EXPLAIN ANY UNCLEAR ONES.

Before analysing, check and report: row count, missing values, duplicates, and anything that looks wrong.

Then answer: BUSINESS QUESTION.

Compute everything with code, not by reading the table. Show me one chart that answers the question, the table behind it, and the code you ran. Tell me how confident I should be and what would change the answer.
Nothing is sent anywhere. It copies to your clipboard.

Alternatives that also work

If GPT-6 Astra is not an option.

Claude Opus 5Official page →

Anthropic · closed · cost: high

Pick it when you work in Claude. It also runs code on uploaded files and writes up the result well.

Gemini 3.8 FlashOfficial page →

Google · closed · cost: low

Pick it when the data already lives in Google Sheets.

Gemma 4 31BHugging Face →

Google · open weights · cost: free weights; you pay for the hardware · 9.1M HF downloads/mo

Pick it when the data cannot leave your machine. Apache 2.0, fits a 24GB GPU at 4-bit, and scores 89.2 on AIME 2026.

Watch out for

Doing this again and again? For weekly reporting and text-to-SQL, use Gemini 3.8 Flash. A reporting pipeline turns questions into queries and query results into a short narrative, many times a week. Gemini 3.8 Flash is Google's "most intelligent workhorse", sits in the low cost tier, and ranks level with Opus 5 on LMArena.

Sources

Checked . Models change monthly; we re-check this page when they do.

Picking the model is the easy part.

Wiring it into a workflow that runs every week, with evals, fallbacks and a cost you can predict, is the work. Fifteen minutes, no deck.

Book fifteen minutes →