Base64 Encoder and Decoder

How to Base64 encode and decode online

1. Encode text: type or paste it. Auto-detect encodes anything that is not valid Base64; choose Encode to force it. Text is converted to UTF-8 bytes first, so emoji, Hindi, Chinese and accented letters work.
2. Decode Base64: paste the string, a Base64URL value from a JWT or URL, or a whole data: URI. Whitespace, line breaks and missing = padding are fine.
3. Decode to a file: if the bytes are an image, a preview appears; PDFs, ZIPs and other binary data get a Download decoded file button with the right extension.
4. Encode an image or file: switch to File or image and drop it in to get a data URI, raw Base64, an <img> tag and a CSS background-image line.

How Base64 works, byte by byte

Base64 takes the input 3 bytes (24 bits) at a time and splits them into four 6-bit groups. Each 6-bit value (0–63) is looked up in the alphabet A–Z a–z 0–9 + /. Encoding Man:

When the last group is short, zero bits fill it and = marks the missing bytes: Ma → TWE=, M → TQ==. That is why Base64 output is always a multiple of four characters and about 33% larger than the input. Press Show bit-level breakdown in the tool to see this for your own input.

Base64 vs Base64URL

Encoding JSON, XML, HTML, CSS, CSV, SVG and other text

Any text format is encoded the same way: its UTF-8 bytes become Base64. What changes is where the result goes, so the Output selector can wrap it in a data URI with the right MIME type:

Decoding works the other way round: paste the Base64 and the tool detects whether the bytes are JSON (offering Pretty-print JSON), XML, HTML, SVG, an image, a PDF or another binary format. To tidy decoded markup, use the XML formatter or the HTML formatter.

The Unicode trap in JavaScript

The browser’s btoa("é") works but btoa("😀") throws “The string to be encoded contains characters outside of the Latin1 range”, because btoa maps each character to one byte. Convert to UTF-8 bytes first — this tool does it for you:

// Encode any string (UTF-8)
const b64 = btoa(String.fromCharCode(...new TextEncoder().encode(str)));
// Decode back to a string
const text = new TextDecoder().decode(Uint8Array.from(atob(b64), c => c.charCodeAt(0)));
// Modern browsers and Node 22+: Uint8Array.prototype.toBase64 / Uint8Array.fromBase64

Base64 in other languages

Python

import base64
base64.b64encode("Café ☕".encode()).decode()      # 'Q2Fmw6kg4piV'
base64.b64decode("Q2Fmw6kg4piV").decode()          # 'Café ☕'
base64.urlsafe_b64encode(b"\xfb\xff").rstrip(b"=")  # b'-_8'

Bash / Linux / macOS

printf '%s' 'user:pass' | base64          # dXNlcjpwYXNz
echo 'dXNlcjpwYXNz' | base64 --decode     # macOS: base64 -D on older versions
base64 -w 0 image.png > image.b64         # GNU: no line wrapping
openssl base64 -A -in image.png           # portable alternative

Java

Base64.getEncoder().encodeToString("Café ☕".getBytes(StandardCharsets.UTF_8));
new String(Base64.getDecoder().decode("Q2Fmw6kg4piV"), StandardCharsets.UTF_8);
Base64.getUrlEncoder().withoutPadding().encodeToString(bytes);   // Base64URL
Base64.getMimeDecoder().decode(wrappedText);                      // ignores line breaks

Go

base64.StdEncoding.EncodeToString([]byte("Café ☕"))
base64.RawURLEncoding.EncodeToString(b)   // Base64URL, no padding
data, err := base64.StdEncoding.DecodeString("Q2Fmw6kg4piV")

PHP

base64_encode("Café ☕");                              // Q2Fmw6kg4piV
base64_decode("Q2Fmw6kg4piV", true);                  // strict: false on invalid input
rtrim(strtr(base64_encode($bytes), '+/', '-_'), '='); // Base64URL

Common questions