Base64 Encoder / Decoder

The Base64 Encoder / Decoder converts plain text to Base64 and back with full UTF-8 support, which matters more than it sounds: JavaScript's built-in atob and btoa choke on emoji, accented letters, and most non-Latin scripts, while this tool encodes multi-byte characters correctly and decodes them without corruption. The Encode tab takes any text and produces standard Base64; the Decode tab reverses it and validates both the character alphabet and the padding, reporting errors like Invalid Base64 — length is not a multiple of 4 instead of guessing. A URL-safe variant switch swaps + and / for - and _, strips trailing padding on encode, and restores both on decode — the exact behavior you need when embedding tokens inside URLs or JWT-style payloads. A Use output as input button feeds the current result into the opposite tab, so you can instantly round-trip a value and confirm it survives encoding. Live stats show the mode plus input and output byte counts, and results download as encoded.b64 or decoded.txt. For decoding a base64 string from an email header, or encoding credentials for a test request, this is the fastest path.

1. 1

   Choose the Encode or Decode tab for the direction you need.

2. 2

   Type or paste your text (or Base64 string) into the input textarea.

3. 3

   Flip the URL-safe variant switch if the value must survive inside a URL or token.

4. 4

   Press Use output as input to feed the current result into the opposite tab for a round-trip check.

5. 5

   Copy the result or download it as encoded.b64 / decoded.txt from the result box.

• Encoding credentials for an Authorization header in a quick API test.
• Decoding a base64 string pasted from an email attachment or a JWT payload section.
• Preparing URL-safe tokens for password-reset links that pass through email clients.
• Embedding small encoded data as a data URI in HTML.
• Decoding base64 log fragments from a mobile crash report.
• Verifying that a round trip through your own encoder preserves UTF-8 characters exactly.

• Never treat Base64 as encryption — it is reversible encoding and offers zero confidentiality.
• Decode JWT payload segments with URL-safe mode on, since segments almost always omit = padding.
• Use the URL-safe variant whenever the string will travel through URLs, query strings, or path segments.
• Remember that standard Base64 output is not whitespace-tolerant everywhere; the decoder here strips whitespace, but strict endpoints may not.
• Test emoji and accented text when validating an encoder pipeline — ASCII-only tests hide UTF-8 bugs.
• Read the decoder's specific errors (bad alphabet character, wrong length) instead of guessing why output is empty.

• Keep the URL-safe switch on while decoding JWT payloads — they almost always omit padding.
• Use Use output as input to prove your pipeline round-trips before trusting it in production code.
• The Output bytes stat is a fast sanity check: decoded ASCII text should roughly match its character count in bytes.
• If decoding produces garbled accents, the source was likely encoded from Latin-1, not UTF-8.