Base64 Encode/Decode

Base64 converts arbitrary binary data (or text) into a string made up only of letters, numbers, +, /, and = — characters that are safe to embed in places designed for plain text, like JSON fields, URLs, email attachments, HTTP headers, or config files. It's not encryption or compression; it's purely a safe-transport format, and anyone can decode it back to the original instantly.

The browser's built-in btoa() function only accepts characters in the 0–255 byte range (Latin-1). Korean characters, emoji, and most non-Latin scripts use multi-byte UTF-8 sequences that fall outside that range, so btoa() either throws an "InvalidCharacterError" or silently mangles the text. This tool avoids that entirely by first converting your text to raw UTF-8 bytes with TextEncoder, encoding those bytes to Base64, and reversing the exact same steps on decode with TextDecoder — so Korean, emoji, and accented characters round-trip perfectly.

Base64 encodes data in 3-byte chunks that become 4 output characters. When the original data isn't a clean multiple of 3 bytes, '=' characters are added at the end to pad the final group to 4 characters — one '=' if 2 bytes were left over, two '=' if 1 byte was left over. Their presence (or absence) is a normal, expected part of valid Base64 — not an error.

It depends entirely on whether the tool sends your input to a server. This tool never does — every conversion runs locally in your browser using built-in JavaScript APIs (TextEncoder, TextDecoder, btoa, atob), with no network request involved. That makes it safe to decode sensitive values like API keys, JWT segments, or Basic Auth headers, which you should never paste into a tool that proxies input through a backend you don't control.