Base64 Encode / Decode
Online Base64 encoder and decoder — supports Chinese, Unicode and special characters
💡 Supports Chinese, Unicode and special characters · Base64 output is ~33% larger than input · All processing is done locally in your browser
About This Tool
Base64 encoding is a fundamental technique that converts binary data into ASCII text, making it safe for transmission over text-based protocols like HTTP, email, and JSON. This free online Base64 encoder/decoder handles everything from short strings to complex data with full Unicode support.
Developers use Base64 encoding for embedding images as data URIs in HTML and CSS, encoding binary data in JSON payloads, handling JWT token parts (header and payload are Base64url-encoded), and storing binary content in text-based databases. The decoder mode is equally useful for inspecting encoded data or debugging API responses.
This tool fully supports Chinese characters, emoji, and Unicode — things that many basic Base64 tools get wrong. All processing happens in your browser with zero server uploads.
How to Use
- Paste the text or Base64 string into the input box.
- The tool auto-detects whether to encode or decode, or switch modes manually.
- Unicode, Chinese and special characters are supported in both directions.
- Click the output to copy the converted string.
Tip: Encoding an encoded string doubles it up — if your output still looks like Base64, you probably encoded twice.
Common Mistakes
1. Treating Base64 as encryption
Base64 is an encoding, not encryption. It provides zero confidentiality — anyone can decode it instantly. Never use it to protect secrets.
2. Broken whitespace
Line breaks and spaces inside a Base64 string make strict decoders fail. Real-world PEM files include newlines every 64 characters and must be stripped first.
3. Corrupted Unicode
The browser btoa/atob functions only handle Latin1 characters. Naive snippets corrupt Chinese and emoji — this tool routes through UTF-8 properly.
4. Missing padding
Strict libraries require the trailing = padding characters, and URL-safe variants swap the plus and slash symbols. Mixing the two alphabets silently breaks decoding.