Decode Base64 strings into readable UTF8 text. Inspect encoded messages and text fields without treating Base64 as encryption.
How to use
Paste the Base64 value into the input. The result appears as text; whitespace in the encoded input is ignored. Correct invalid characters if decoding fails.
Example
SGVsbG8= decodes to Hello. The equals sign is padding, not part of the decoded word.
FAQ
Can I decode an image here?
No. This decoder requires valid UTF8 bytes and does not export binary files. Use the original file format for images or archives.
Related tools
A practical task
For an API sample, copy only the encoded field, not surrounding JSON quotes. Start with a known value such as VGVzdA==, which reads Test. If the value is actually a binary file, readable text decoding is the wrong destination even when its alphabet looks valid.
Another useful question
Why do decoded words look wrong?
Base64 restores bytes, not their original encoding label. This decoder expects UTF8. Bytes from an older character encoding need conversion with the correct source encoding.
Details for interpreting the result
| Detail | Explanation |
|---|---|
| UTF8 bytes | UTF8 encodes characters as bytes. Basic Latin letters use one byte each, while many accented letters and other scripts use more. A byte count therefore differs from a visible character count. |
| Base64 size | Base64 normally stores three input bytes as four visible characters, with padding for incomplete final groups. Encoded text is therefore usually longer than its source bytes; it is not compression. |
| Base64 padding | An equals sign at the end of Base64 fills a final group. It is not part of the decoded message. Do not remove characters from the middle of the value to force a damaged string through a decoder. |
| Standard and URL Base64 | JWTs use a URL friendly Base64 variant. Its alphabet differs from standard Base64 for two characters, and padding is often omitted. A token section and an ordinary Base64 string are not interchangeable without conversion. |
| A complete byte | A byte holds a value from 0 through 255. Binary writes that value with eight digits; hexadecimal writes it with two. Leading zeros are part of that fixed width representation and must stay in place. |
| Encoding and encryption | Encoding changes how data is represented so another system can read it. It does not require a secret key. Encryption protects information with a key, while hashing produces a digest for a different purpose. |
| Similar looking text | Two strings can look identical but use different Unicode sequences. An accented letter can be stored directly or as a letter plus an accent. Visual similarity alone does not establish byte equality. |