Is my Base64 uploaded anywhere?

No. Decoding runs entirely in your browser; nothing is sent to a server.

Does this accept URL-safe Base64?

Yes — the - and _ characters used by URL-safe Base64 (common in JWTs) are normalized automatically before decoding.

What happens if the input isn't valid Base64?

The tool reports that the text contains characters outside the Base64 alphabet or has invalid length instead of guessing at a result.

Why do I get a UTF-8 error on decode?

That means the decoded bytes aren't valid UTF-8 text — likely because the Base64 actually encodes binary data (an image or file) rather than text.

Base64 Decode

Decode Base64 to text or check it for validity, with correct UTF-8 handling and automatic support for the URL-safe alphabet. Nothing is uploaded to a server.

Loading tool…

Worked examples

A standard Base64 string with padding
The everyday case: decoding a Base64 value copied from an API response, JWT segment or config file back to readable text.
Multi-byte UTF-8 characters
Correct UTF-8 decoding matters — treating the bytes as Latin-1 or ASCII instead would corrupt accented letters and emoji.
URL-safe Base64 without padding
URL-safe Base64 (from a JWT or a query parameter) uses - and _ in place of + and / and omits padding; this tool normalizes both forms automatically.

What this tool does

This tool reverses Base64 encoding: it takes a Base64 string and decodes it back to the original UTF-8 text, entirely in your browser. It accepts both the standard alphabet (+, /, = padding) and the URL-safe alphabet (-, _, no padding) automatically, so you don't need to know which variant you're looking at before pasting it in.

When you need it

Reading a Base64 value copied from an API response, a JWT segment, a config file, or an Authorization: Basic header back to plain text.
Checking what's actually inside a Base64 string before trusting it — a quick way to confirm a token or payload contains what you expect.
Debugging a webhook or integration that sends Base64-encoded fields, where you need to see the decoded value to diagnose a problem.

Rules the decoder applies

Both Base64 alphabets are accepted without you specifying which one it is. The - and _ characters from URL-safe Base64 are normalized to + and / internally before decoding, and missing padding is added automatically.
Whitespace is stripped before decoding, since Base64 pasted from an email or a line-wrapped source sometimes has line breaks inserted at fixed intervals.
Input is validated against the Base64 character set first. Text containing characters outside the alphabet (or with an invalid length) is rejected with a clear error rather than silently producing garbage output.
Decoded bytes are interpreted as UTF-8 text. If the bytes aren't valid UTF-8 — which happens when the Base64 actually encodes binary data like an image rather than text — the tool reports that explicitly instead of showing corrupted characters.

Why you might see a UTF-8 error

Base64 can encode any bytes at all, not just text — images, PDFs, encryption keys and other binary formats are all commonly Base64-encoded for the same "safe to put in plain text" reason. When you decode Base64 that actually holds binary data, the resulting bytes usually aren't valid UTF-8, so this tool reports the mismatch rather than displaying mojibake. That error is itself useful information: it tells you the Base64 string you're looking at is binary data, not encoded text, and you'd need a different tool (one that can output a downloadable file) to work with it.

Things to check after decoding

If the decoded text looks like it's missing characters or has odd substitutions, double-check you copied the entire Base64 string — a single dropped character shifts every subsequent group of four and corrupts the rest of the decode.
Remember decoding isn't decryption: if the original value was Base64-encoded and encrypted, decoding only undoes the encoding step, and you'll see encrypted bytes (or a UTF-8 error) rather than the original plaintext.

Limits

Input is capped at 2 MB of Base64 text. This tool outputs text — if you're decoding a Base64 string that represents an image or other binary file, you'll need a tool designed to reconstruct and download that file type instead.

Frequently asked questions

Is my Base64 uploaded anywhere?: No. Decoding runs entirely in your browser; nothing is sent to a server.
Does this accept URL-safe Base64?: Yes — the - and _ characters used by URL-safe Base64 (common in JWTs) are normalized automatically before decoding.
What happens if the input isn't valid Base64?: The tool reports that the text contains characters outside the Base64 alphabet or has invalid length instead of guessing at a result.
Why do I get a UTF-8 error on decode?: That means the decoded bytes aren't valid UTF-8 text — likely because the Base64 actually encodes binary data (an image or file) rather than text.