Why does native JavaScript btoa() throw an error on Unicode emojis and non-Latin characters?

The standard browser window.btoa() function treats characters as binary 8-bit strings (Latin1 range 0x00–0xFF). When given characters outside this range (such as emojis or Arabic/Devanagari letters), btoa() throws an "InvalidCharacterError". Our tool uses TextEncoder() to safely convert Unicode strings into raw UTF-8 byte streams before Base64 encoding.

What is the difference between Base64 and URL-Safe Base64 (Base64URL)?

Standard Base64 uses the characters "+" and "/" and includes "=" padding. In URLs, query strings, and JWT headers, these characters have special delimiter meanings. Base64URL (RFC 4648 §5) replaces "+" with "-" and "/" with "_", and strips trailing "=" padding for safe URL embedding.

Why does Base64 encoding increase data size by approximately 33%?

Base64 represents 6 bits of binary data per ASCII character. Because standard bytes contain 8 bits, every 3 input bytes (24 bits) require 4 Base64 characters (24 bits), resulting in a mathematical size expansion of 4/3 (+33.33%).

Tools/Base64

Unicode Base64 & Base64URL Studio

Encode Unicode text to standard Base64, URL-Safe Base64URL (for JWTs), and Data URIs, and decode byte streams back to readable UTF-8 text with zero corruption.

73 BRaw UTF-8 Byte Size

100 CharsBase64 String Length

+37%Base64 Expansion Ratio

standardActive Encoder Mode

Encoding Specification / Target:

Quick Presets:

Font Size (16px):

UTF-8 Unicode Text (To Encode)Live Input Buffer

Hello World 🚀! UTF-8 Multi-Byte String with Unicode 16.0 & Symbols ★

Base64 Encoded Output

SGVsbG8gV29ybGQg8J+agCEgVVRGLTggTXVsdGktQnl0ZSBTdHJpbmcgd2l0aCBVbmljb2RlIDE2LjAgJiBTeW1ib2xzIOKYhQ==

24-Bit Quantum Chunk Anatomy (3 Bytes → 4 Base64 Characters)

Input Chars	8-Bit Raw Bytes (Hex / Decimal)	6-Bit Base64 Output
Hel	0x48 (72), 0x65 (101), 0x6C (108)	SGVs
lo	0x6C (108), 0x6F (111), 0x20 (32)	bG8g
Wor	0x57 (87), 0x6F (111), 0x72 (114)	V29y
ld	0x6C (108), 0x64 (100), 0x20 (32)	bGQg
�	0xF0 (240), 0x9F (159), 0x9A (154)	8J+a
�!	0x80 (128), 0x21 (33), 0x20 (32)	gCEg

UTF-8 Safe Base64 Encoding Ready to Run

Standard library base64 encoding and decoding.

1import base64

3text = "Hello, Unicode! 🌍 (Urdu: سلام)"

5# 1. Base64 Encode (UTF-8 bytes to ASCII base64)

6encoded = base64.b64encode(text.encode('utf-8')).decode('ascii')

7print("Encoded:", encoded)

9# 2. Base64 Decode back to Unicode String

10decoded = base64.b64decode(encoded).decode('utf-8')

11print("Decoded:", decoded)

Python 3Zero external runtime dependencies • Standard Library

UTF-8 & Unicode 16.0 Compatible

The Mathematics & Bit Architecture of Base64 Encoding (RFC 4648)

Base64 is a binary-to-text encoding scheme that represents raw binary data in an ASCII string format. It transforms 24 bits (3 bytes of 8 bits each) into 4 six-bit groups (6 bits each):

• 6-Bit Alphabet (64 Glyphs): A-Z (0–25), a-z (26–51), 0-9 (52–61), + (62), / (63).

• Padding Characters: If the input length is not a multiple of 3 bytes, = or == padding is appended to fill the 24-bit quantum block.

• Size Overhead: Exactly $4 / 3 = 1.333$ (+33.33% expansion in data transmission).

Standard Base64 vs. URL-Safe Base64 (JWT & OAuth PKCE)

•Standard (RFC 4648 §4): Uses + and / with = padding. Used for email attachments (MIME) and file transfers.
•Base64URL (RFC 4648 §5): Replaces + with - and / with _ without padding. Essential for JSON Web Tokens (JWT) and URL query safety.

Frequently Asked Questions (FAQs)

Why does native JavaScript btoa() throw an error on Unicode characters?+

The standard browser window.btoa() function treats characters as binary 8-bit Latin1 strings (0x00–0xFF). Characters outside this range (emojis, Urdu, Arabic, Hindi) cause btoa() to fail with an InvalidCharacterError. Our tool uses TextEncoder() to convert strings into raw UTF-8 bytes before encoding.

How do I use Base64 in JSON Web Tokens (JWT)?+

JWT tokens require the "URL-Safe Base64URL" format (without "+" or "/" and without "=" padding). Select the "URL-Safe Base64URL" mode for full JWT compatibility.

Can I decode Data URIs directly?+

Yes! Paste the full "data:text/plain;charset=utf-8;base64,..." string and select "Decode". Our parser automatically strips the header and extracts the underlying text.