KOI8-R is a single-byte character encoding for Russian Cyrillic defined in RFC 1489. It was the dominant encoding for Russian email and Usenet before UTF-8, and is still seen in legacy files, headers, and protocols.

Why is KOI8-R laid out the way it is?

Its clever design places each Cyrillic letter so that clearing the high bit yields a phonetically similar Latin letter. So if the 8th bit is stripped by an old 7-bit system, КОИ8 text degrades into a roughly readable Latin transliteration rather than gibberish.

How many bytes per character?

KOI8-R is a single-byte encoding, so each character is exactly one byte from 0 to 255. Bytes 0x00 to 0x7F match ASCII, and the upper half 0x80 to 0xFF holds Cyrillic letters and box-drawing characters.

What if my text has characters KOI8-R cannot store?

Only ASCII and Russian Cyrillic (plus some box-drawing glyphs) fit in KOI8-R. Ukrainian-specific letters, emoji, or other scripts have no byte and are marked with ?? when encoding so you can see what was dropped.

Is the conversion done locally?

Yes. The byte table is built in your browser from the native decoder and applied locally for both directions. Nothing is sent to a server.

KOI8-R Encoder/Decoder

KOI8-R is the classic single-byte encoding for Russian Cyrillic, standardised in RFC 1489 and ubiquitous in pre-Unicode Russian email, news, and Unix systems. This tool converts Cyrillic text to KOI8-R hex bytes and decodes KOI8-R bytes back to text.

How it works

KOI8-R maps every character to exactly one byte. The lower half 0x00–0x7F is plain ASCII; the upper half 0x80–0xFF carries Cyrillic letters and some box-drawing symbols. Its most famous property is the deliberate ordering of the Cyrillic block: each letter sits at the code point whose low 7 bits spell a phonetically related Latin letter, so stripping the high bit turns Cyrillic text into a rough Latin transliteration rather than noise.

The tool builds the exact byte-to-character table in your browser using the platform’s native KOI8-R decoder, then inverts it for encoding. Decoding feeds your hex bytes straight through that same decoder.

Example and notes

"Привет, мир" encodes to the KOI8-R bytes f0 d2 c9 d7 c5 d4 2c 20 cd c9 d2 — one byte per Cyrillic letter, with the comma and space staying as their ASCII 0x2C and 0x20.
KOI8-R covers Russian but not all Cyrillic; some Ukrainian letters live in the related KOI8-U variant and will be flagged as unmapped here.
When decoding, tokens that are not valid 0–255 hex values are skipped and the tool tells you so.