Base64 Encoder & Decoder

Type or paste text on either side. The other side updates as you type. UTF-8 safe by default — emoji, CJK, accents all roundtrip correctly.

What this does

• Two-way live conversion. Type in either pane; the other auto-updates.
• UTF-8 safe. Unlike a plain btoa() call, this tool encodes the text as UTF-8 bytes first, so non-Latin characters survive the roundtrip.
• URL-safe variant. Tick "URL-safe" to swap + → - and / → _, the form used in JWT and URL query strings.
• Strip padding. Drop trailing = for compact output (the decoder accepts both forms).
• Line wrap. Insert line breaks every 76 characters, MIME-style. Useful for email and PEM-style text.

Why client-side

Most "Base64 encoder" sites upload your text to their server, encode it there, and return the result. That's a privacy and bandwidth tradeoff for nothing — every modern browser has built-in base64 functions.

This tool uses btoa() (encode) and atob() (decode) plus the TextEncoder/TextDecoder APIs for UTF-8. Total JS for the page is under 10 KB. Nothing leaves your browser.

Common use cases

• HTTP Basic Auth. The Authorization: Basic header takes base64(user:pass).
• Data URIs. Inline images, fonts, or PDFs in HTML/CSS use data:image/png;base64,.... See the file converter.
• Email attachments. MIME wraps binary attachments as base64.
• Configuration values. Kubernetes secrets, env vars, and many config formats expect base64-encoded credentials or certs.
• Storing binary in JSON. JSON has no binary type — base64 is the conventional escape.
• JWT. The three parts of a JWT are base64url-encoded JSON. See our JWT decoder.

Programmatic equivalents

// JavaScript / Node
btoa(unescape(encodeURIComponent(text)));        // encode (UTF-8 safe)
decodeURIComponent(escape(atob(base64)));        // decode

// Python
import base64
base64.b64encode(text.encode("utf-8")).decode()
base64.b64decode(b64_string).decode("utf-8")

// Bash
echo -n "hello" | base64
echo -n "aGVsbG8=" | base64 -d

// Go
import "encoding/base64"
base64.StdEncoding.EncodeToString([]byte(s))

Common pitfalls

• btoa alone breaks on Unicode. btoa("café") throws. Always UTF-8-encode first.
• URL-safe is not interchangeable. A URL-safe base64 string fails strict decoders unless you replace -/_ back to +// first.
• Padding requirements vary. Some decoders require trailing =, others reject it. RFC 4648 says padding is optional for url-safe encoding.
• Don't use base64 as encryption. It's encoding, not cryptography. Anyone can decode it.