Markdown table to Excel. Straight from ChatGPT.
Paste a table from ChatGPT, Claude, Gemini or Copilot and get a real .xlsx workbook — no asterisks, no broken columns.
Your table preview will appear here
How to use Markdown Table to Excel
Paste the AI's answer
Copy the whole reply — surrounding prose and all. Parsefy finds the markdown tables inside it and ignores the rest.
Check the parsed tables
Bold markers and links are flattened to plain text, numbers are detected as numbers, and each table is previewed before you commit to a download.
Download or paste into Docs
Take a real .xlsx workbook with one sheet per table, or use Copy for Google Docs to paste a formatted table straight into Docs or Word Online.
Why use Parsefy Markdown Table to Excel?
Your data never leaves your device
Everything runs in your browser tab — there is no upload step and no server to store anything. That makes it safe for API responses, database dumps and production data you could not paste into a hosted converter.
Every table, on its own sheet
A long AI answer often contains several tables. Each one becomes a separate sheet, named after the heading above it, instead of being mashed into a single grid you have to split by hand.
No more literal **asterisks**
Pasting markdown into Excel leaves the formatting syntax visible as text. Parsefy resolves bold, italics, inline code and links to the words underneath, so cells contain the value and nothing else.
No sign-up, no file-size tier
There is no account, no email wall and no paid plan gating bigger files. The only real limit is how much memory your own browser can spare.
Markdown Table to Excel — frequently asked questions
Why does pasting a ChatGPT table into Excel break?
Because the answer is markdown, not a table. Excel receives one long run of text per line, so the pipe characters end up inside cells, the |---|---| divider becomes a data row, and formatting markers like **Total** stay visible as literal asterisks. Parsefy parses the markdown first and builds a real spreadsheet from the result.
Is my prompt or data sent anywhere?
No. The parsing and the workbook are both built in JavaScript inside your browser tab, so pasted text never travels over the network. That matters here more than usual, since AI answers often contain work data you could not paste into a hosted converter.
What happens if the answer contains several tables?
All of them are converted. Each table becomes its own sheet, named after the nearest heading above it, and you can click through the tabs to preview each one before downloading.
Will numbers arrive as numbers or as text?
Plain numeric cells are written as real numbers so you can sum a column immediately. Values that only look numeric are deliberately kept as text — leading zeros such as 007, IDs longer than 15 digits, and anything with a currency symbol, percent sign or thousands separator, all of which would otherwise be silently altered.
Which AI tools does this work with?
Any of them. ChatGPT, Claude, Gemini, DeepSeek, Copilot and Perplexity all emit standard GitHub-flavoured markdown tables, and the parser follows that spec. It works equally well on a table copied out of a README or a docs page.
What if the answer is messy — fenced, ragged, or cut off?
That is the normal case, and it is repaired rather than rejected. A ```markdown fence wrapped around the answer is removed, prose before and after the table is ignored, rows with too few columns are padded and rows with too many are trimmed, and a divider row that disagrees with the header defers to the header. If the answer was cut off mid-table, the complete rows are recovered and you are told how many. Every repair is listed above the preview, so nothing is changed without you seeing it.
Is there a limit on how big the table can be?
An Excel worksheet holds 1,048,576 rows and 16,384 columns, and a single cell holds 32,767 characters. Parsefy checks your data against those limits and warns you before it builds a file Excel would refuse. In practice you may meet a different ceiling first: because the workbook is assembled in your browser, very large sheets may hit your browser's memory limits before they hit Excel's own limits.