Turbo Large Data Splitter

Turbo Large Data Splitter is a Windows tool that splits CSV, Excel and other large data files into smaller ones and merges them back together, without the row limits and memory problems of opening huge files in a spreadsheet. It reads CSV, TSV, TXT, XLSX, XLS, JSON, XML and SQL dump files, and can convert formats while splitting — for example turning one giant CSV into several smaller XLSX workbooks. Every file streams through in a single pass, so the practical limit is free disk space rather than RAM, and files over 100 GB work.

A single Excel worksheet stops at roughly 1,048,576 rows, and long before that a multi-gigabyte export can freeze a machine or refuse to open. Seven split methods — rows, file size, column value, date, alphabet range, duplicates and unique values — turn one enormous export into a set of files sized for whatever comes next, and the Merge tab reassembles parts back into one file with a single header.

Input and output are independent settings, so CSV to XLSX, XLSX to CSV, JSON to CSV and similar pairs all work in the same pass as the split or merge. Leave the output on "(same as input)" for a plain split. SQL dump files are read line by line, so each line is treated as one row; SQL is an input format only.

Each method needs a single setting: rows per file, target MB, the column to split on or the date granularity. The header row is kept on every part.

• CSV to XLSX. Turn a giant CSV export into several Excel workbooks sized under the worksheet row limit, ready to open and share.
• XLSX to CSV. Convert a workbook, including a chosen sheet from a multi-sheet file, into lightweight CSV parts for import elsewhere.
• JSON to CSV. Flatten a large JSON array into tabular CSV files that open in any spreadsheet, split into manageable pieces along the way.
• SQL dump to rows. Each line of a line-oriented dump becomes one record that can be written out as CSV, JSON, XML and more.

• Add your file. Browse, drag it onto the window or pick it from Recent Files. The input format is detected from the extension.
• Pick input and output formats. Keep "(same as input)" for a plain split, or choose CSV, TSV, TXT, XLSX, JSON or XML to convert while splitting.
• Choose a split method and fill in its single setting.
• Set output and compression. Choose the output folder, optionally compress each part or bundle everything into one archive, and adjust the advanced CSV settings only if needed.
• Run Preview. It reports the row count, detected columns, the estimated number and size of output files, and a note when a conversion is set.
• Start Split. Watch progress, live speed and ETA in the status bar, then open the output folder.

Reading and splitting happen in one continuous, ordered pass, which keeps row order correct and makes resuming possible. Because files are streamed rather than loaded, the practical limit is free disk space, and 100 GB+ files are supported. Excel files are read with a high-speed engine, which keeps large XLSX-to-CSV and XLSX-to-XLSX splits quick. Only compressing the finished parts is parallel, on up to 128 workers, which speeds up jobs that produce many ZIP or GZIP files.

• Make a huge export open in Excel. Choose Number of Rows at 1,000,000 and set the output format to XLSX; a 12-million-row CSV becomes twelve workbooks that each open cleanly with the header row.
• Distribute data by region. Pick Column Value and the Region column for one file per region. Apply a simple filter for the current quarter and sort so each file arrives in a sensible order.
• Convert a JSON feed to spreadsheets. Load the JSON, set the output to XLSX, choose Number of Rows and split; a nested export becomes tabular workbooks your team can open immediately.
• Archive transactions by month. Choose Date Column, set the granularity to month and point it at the order-date column; twelve monthly files come out in one pass.
• Reassemble split parts. On the Merge tab add the files in order, choose where to save and click Merge Files for one file with a single header.

• Row-based splitting is capped at 1,000,000 rows per output file, which keeps each part within Excel’s worksheet ceiling.
• Legacy .xls is read by loading the whole workbook into memory, which is fine because .xls itself tops out at 65,536 rows.
• Previewing and counting rows in an XLSX requires a full read of the sheet, so it is not instant on very large workbooks.
• Duplicate and Unique splitting keep a hash of every row seen in memory, which works well to tens of millions of rows; column or alphabet splitting suits extreme cardinalities better.
• SQL dumps are split by treating each physical line as one row. It is not a full SQL parser, and SQL is an input only.

All splitting, merging, sorting and filtering happen on your own computer. No files are uploaded, scanned remotely or shared with third parties, and there is no server-imposed file-size cap. Internet is used only for license validation, and the request contains no file names or file content. Suitable for financial records, customer data and other confidential datasets.

Turbo Large Data Splitter is distributed through Gumroad and activated with the email and license key from your purchase receipt. Buy a license with the Buy Now button on this page, run the installer or packaged executable on Windows 10 or 11, and enter the email and key at first launch; no browser, server or online account is needed. After activation the app opens straight to the Split tab: drag a file onto the window, pick a method, run Preview and click Start Split.