Comparisons
Bigger than a spreadsheet.
Faster than a script.
Every row of a 10 GB CSV, with no code and no upload.
vs Excel50× more rows52,720,416 rows, every one visible
10 GB CSV2.45 sto open and index, on a laptop
vs pandas57.3× fastertypical, CSV to answer
vs DuckDB4.3× fastertypical, CSV to answer
If you open CSVs in an app
Excel, Google Sheets, Modern CSV, EmEditor
csvlite52,720,416every row
Google Sheets1,333,3332.5%
Excel worksheet1,048,5762.0%
Limits from Microsoft and Google, checked 29 September 2026.
| Tool | What it is | Edits the file | Big files | Runs on |
|---|---|---|---|---|
| csvlite | Data viewer and analyzer | No, read-only by design | Every row; 52.7M rows open in 2.45 s | Windows, Linux |
| Excel ↗ | Spreadsheet | Yes | 1,048,576 rows per worksheet | Windows, macOS, web |
| Google Sheets ↗ | Online spreadsheet | Yes | 20 million cells; the file is uploaded | Web |
| Modern CSV ↗ | CSV editor | Yes | Read-only mode for large files | Windows, macOS, Linux |
| EmEditor ↗ | Text editor | Yes | Files up to 16 TB | Windows |
If you write code
pandas, Polars, DuckDB, VisiData
csvlite1.4 s
DuckDB9.9 s
Polars5.4 s
pandas29 s
vs pandas57.3×typical speed-up · faster on 27 of 27The code, and more data →vs Polars2.2×typical speed-up · faster on 34 of 35The code, and more data →vs DuckDB4.3×typical speed-up · faster on 30 of 36The code, and more data →DuckDB & VisiDataSQL or keys?When a query language or a terminal is the better fitCompare the workflows →
Read-onlyYour source file is never modified.
Air-gap readyNo network code. Nothing is uploaded, nothing phones home.
Every rowFilter, sort, profile and pivot the whole file.
Other products’ facts are from their own documentation, checked 29 September 2026. Speed: measured comparison · csvlite benchmarks