Choose a guide for your problem
How to fix Base64 decoding errors
Diagnose which layer failed: Base64 syntax, text interpretation, or file restoration. This page is the diagnostic starting point; the other guides explain each layer in depth.
Base64 encoding Korean text and emoji with UTF-8
Compare identical text encoded as different bytes. Select a character encoding from knowledge of the source, not from the language of the website.
Base64 vs Base64URL: differences, padding and decoding examples
Follow a Base64 value through a query string. This guide isolates transport damage from errors in the original value.
Base64 padding explained: what = and == mean
Work out padding from byte groups and check malformed endings. Character encoding and URL transport are covered separately.
JavaScript Base64 with Unicode: fix btoa() errors using UTF-8
Implement Unicode conversion in JavaScript and inspect cases where a successful btoa call still gives the wrong byte encoding.
Verification method
These fixed, public fixtures run against the same file codec and character codecs used by the converter. The JavaScript cases additionally check browser-native btoa, TextEncoder, URLSearchParams and SHA-256. Expected outputs are published before any script runs; the button shows actual results in your browser. No input is uploaded.
Checked on 2026-10-21. This is a correctness check with small fixtures and one 100,000-character case, not a speed benchmark or a guarantee for every file. The four-byte file is synthetic binary data, not a PNG or PDF. SHA-256 equality checks byte preservation; it does not establish that a file is safe to open. Results can depend on browser support.