A 100 KB CSV test file

About this file

A 100 KB test file in CSV format, sized to exactly 102,400 bytes. Free to download with no signup, for verifying upload limits and file-handling behaviour.

The MD5, SHA-1 and SHA-256 values listed above are the real digests of this file. Recomputing them after download confirms the transfer was not corrupted, and they double as known-good expected values when testing a hashing library.

What this size means in practice

100 KB sits at the boundary between "trivially small" and "realistic payload". It is close to the total transfer weight of a single lean web page and is a common target size for performance-conscious images.

It clears the sub-1 MB caps many applications impose, but it is still small enough that a naive load-everything-into-memory implementation works fine — so it is not the right size for proving that streaming is necessary. Pair it with 10 MB or larger to see the difference.

Getting a feel for this size

On a 10 Mbps link the transfer takes roughly 82 ms, and about 8 ms at 100 Mbps. Base64-encoded it grows to roughly 133 KB.

For a typical business CSV of about ten columns and roughly 100 bytes per row, this size corresponds to about 1,024 rows — a useful gauge for whether a spreadsheet can open it or a bulk database import finishes in time. It is a rough estimate; actual counts vary with content and column count.

What to check for this format

CSV is plain text and has no magic bytes of its own. You therefore cannot prove a file is CSV by inspecting a signature; validation has to focus instead on character encoding, delimiters, and quoting rules.

• Whether you process row by row instead of loading every line into memory — the classic cause of out-of-memory failures on large CSVs
• Whether quoted values containing embedded newlines or commas are parsed correctly
• Whether encoding detection is correct — CSVs containing Japanese are a frequent source of mojibake