Binary Code Translator

One box takes either kind of input, and the output updates as you type.

1. Type or paste into the Input box. Letters, digits, punctuation, accents and emoji are all accepted.
2. Leave Mode on Auto detect and the tool decodes when the input is only zeros, ones and white space and is at least eight characters long. Otherwise it encodes. Choose Text to code or Code to text to force a direction.
3. Read the result. Each byte is eight digits, with one space between bytes.
4. To decode, paste groups of exactly eight digits, separated by spaces or line breaks, or one unbroken run of digits.
5. Press Copy, Download .txt or Use as input to send the result back in.

Decoding looks for runs of eight digits, so a space inside a byte, such as 0100 1000, breaks it, and groups of seven or four digits are not read at all.

Each character has a number, and that number is written in base 2 using place values of 128, 64, 32, 16, 8, 4, 2 and 1. The letter H is 72, which is 64 + 8, so its byte is 01001000. The letter i is 105, which is 64 + 32 + 8 + 1, giving 01101001. So Hi becomes 01001000 01101001.

Characters beyond plain English use UTF-8, which spends more than one byte on them. The euro sign becomes three bytes, 11100010 10000010 10101100, and é becomes two. The same bytes written in hexadecimal are shorter, which the hex to text converter shows, and the UTF-8 encoder explains how the byte patterns are built.

Two patterns make short messages readable without a tool. Capital letters begin 010 and lower-case letters begin 011, so the two cases differ only in the third bit, worth 32. Digits begin 0011, so 00110101 is the character 5.

Auto detect has one trap. A string such as 10101010 looks like a byte, so it is decoded, and because that value is not valid UTF-8 on its own you see a replacement character. If you really want the typed digits as text, set Mode to Text to code, which gives 00110001 00110000 ... for each digit. Try a decode by hand: 01000001 01001110 01000100 is 65, 78 and 68, which are the letters A, N and D, so the message reads AND. This tool converts characters, not quantities. To write the number 255 in base 2, which is 1111 1111, use the number base converter.

The translator keeps no history and has no file input, so long documents are better handled in a text editor or a script. It always works in 8-bit UTF-8 bytes. It does not produce 7-bit ASCII groups, 16-bit UTF-16 or little-endian variants, and it does not reverse bit order. Binary here is a way of writing text, not security: anyone can paste it back in and read it. It is also unrelated to compiled machine code, which is binary in a different sense.

Bit patterns also describe settings, not only text. When a byte holds eight on/off flags, such as file permissions or hardware status bits, writing it out makes each flag visible: counting from the right as bit 0, the byte 00000101 has bits 0 and 2 set. People use it to check classroom exercises, to hide a short message in a puzzle or escape room, and to see which bits differ when a flag or character looks wrong. For another classic puzzle code try the Morse code translator.

A practical tip for teaching: encode a short word, hide the output, and ask learners to decode it by hand using the place values above. Then paste their answer in as binary to check it. Because every letter is eight digits, mistakes such as a dropped digit are easy to spot, since the group no longer has the right length and the decoder skips it.