Runs in-tab
All encoding and decoding happens in your browser. No requests, no logs, no storage.
3 bytes in.4 chars out.
base64s.com is a Base64 encoder/decoder that runs entirely in your browser tab. Zero uploads. Zero limits. Zero accounts.
01 / features
Local. Fast. Exact.
Built for people who paste JWTs, API keys and binary blobs, and would rather not send them anywhere.
No caps
No upload ceiling and no queue. Throughput is bounded only by your device.
Every option
Standard or URL-safe alphabet, padding on or off, full UTF-8, auto-detect on decode.
02 / uses
Binary over text-only pipes.
Anywhere bytes need to travel through a channel that only understands characters.
Data URIs
Inline small images in HTML and CSS.
data:image/png;base64,iVBOR…JSON APIs
Ship files and signatures as strings.
"blob":"JVBERi0xLjQK…"MIME email
Every attachment is Base64 on the wire.
Content-Transfer-Encoding: base64JWT & IDs
Base64URL keeps tokens URL-safe.
eyJhbGciOiJIUzI1NiJ9.…03 / code
Already in your stdlib.
Pick your runtime. Same bytes, same output.
04 / faq
man base64
Short answers to the questions that come up most.
Is Base64 encryption?
No. It's a reversible encoding. Anyone can decode it. Use AES or similar when you need confidentiality.
Base64 vs Base64URL?
Base64URL (RFC 4648 §5) replaces + / with - _ and usually omits =. Use it in URLs, filenames and JWTs.
What's the trailing "="?
Padding to a multiple of 4 chars. 1 leftover byte → XX==, 2 → XXX=. Most decoders accept unpadded input.
Why is the output bigger?
4 chars per 3 bytes = ~33% overhead. 1 MB in → ~1.37 MB out.
Why is my decode garbled?
Whitespace in the input, std/url alphabet mismatch, or binary (PNG, ZIP) rendered as text. Strip, switch alphabet, or save as a file.
Does anything get uploaded?
No. base64s is built to run 100% client-side. There is no backend for your data.
Running here.
Paste into the box at the top. Encoding and decoding already run in this tab.