What Is Base64 Encoding? A Plain-English Explanation

Base64 is a way of representing binary data — like an image, a file, or raw bytes — using only 64 printable text characters (A-Z, a-z, 0-9, +, /). It shows up constantly in web development: embedding small images directly in CSS, sending file data inside JSON, or encoding credentials in an HTTP Authorization header.

Why not just send the raw bytes?

Many systems — email, URLs, JSON, HTML attributes — were built to carry text, not arbitrary binary data. Raw bytes can contain values that those systems interpret as control characters or break on. Base64 sidesteps this by re-encoding any binary data as a string using only safe, printable characters, at the cost of making the data about 33% larger.

Common places you'll see Base64

• Data URLs — embedding a small image directly in HTML/CSS as data:image/png;base64,...
• JSON Web Tokens (JWTs) — the header and payload are Base64-encoded JSON
• Email attachments — MIME encodes binary files as Base64 to travel through text-based email protocols
• HTTP Basic Authentication — the username:password pair is Base64-encoded (not encrypted) in the request header

How to encode or decode it

A free Base64 Encoder/Decoder handles both directions instantly — paste text or a file to encode it, or paste a Base64 string to decode it back to its original form, all in your browser.

How Base64 actually works

The algorithm groups raw binary data into chunks of 3 bytes (24 bits), then splits each chunk into four 6-bit groups. Since 6 bits gives 64 possible values, each group maps to one of 64 printable characters — hence "Base64." When the input doesn't divide evenly into 3-byte chunks, the output gets padded with one or two = characters at the end, which is why you'll often see a Base64 string ending in = or ==.

Base64 vs URL encoding vs hashing

These three get confused often, but they solve different problems. Base64 makes binary data text-safe and is fully reversible. URL encoding (percent-encoding) escapes specific characters that have special meaning in a URL, like spaces and &, and is also reversible. Hashing (like SHA-256) is one-way — it produces a fixed-length fingerprint of data and can't be reversed back to the original at all. If you need to get the original data back, you want encoding, not hashing.