Wubi code converter (86/98)
Convert Chinese characters to Wubi (五笔字型) full codes: a built-in 527-character 86-version table, single-character or whole-text input, one line per character or compact output, and a 98-version switch. Characters outside the table are labelled instead of guessed.
Runs in your browserEvery computation happens in your browser — your data never leaves this device.
Wubi codes
Codes come from the built-in 86-version table (full codes); characters that are not in the table are labelled "not included" instead of being guessed.
Enter Chinese characters to see their Wubi codes
What this tool does
- Look up the Wubi code of a single character: type 中 and get KHK, type 国 and get LGYI — useful for typing practice, checking textbooks or filling in forms that ask for Wubi codes.
- Convert a whole Chinese sentence: pick "Whole text" to get one line per character (字 编码) or switch to compact mode, which joins the same 字 编码 pairs into a single space-separated line (`中 KHK 国 LGYI ...`) ready to paste into a typing drill.
- Compare the 86 and 98 versions: in 98 mode, characters recorded as differing between the two versions get their 98 code, while every other character falls back to the 86 code and is labelled "按 86 版" so the two versions are never mixed up.
- Check characters that are not covered: rare or variant characters are reported as "not included" and counted, instead of being given a plausible-looking code.
Example
Input
中
Output
中 KHK
Wubi 86 full code. 中 = 口 (K) + 丨 (H) + shape code for a vertical ending in an enclosed character (K).
Frequently asked questions
How many characters are covered, and why do some show "not included"?
The bundled table holds 527 common characters with their Wubi 86 full codes, covering the key-name characters (王土大木工…), common character roots and frequent everyday characters. Rare characters, variant forms, traditional characters and polyphonic readings are not fully covered; anything outside the table is reported as "not included" and counted rather than guessed.
Which encoding rules are used, and do the codes match what my IME outputs?
The tool follows the Wubi 86 rules: key-name characters repeat the key four times (人 WWWW), character roots use key + first + second + last stroke (五 GGHG), and compound characters take roots 1-2-3-last plus a shape code when fewer than four roots are available (不 GII, 中 KHK). Output is the full code; the 1–3 letter short codes your IME shows (e.g. 中 K) are shortened by character frequency from the same full code, so they are related but not the same length.
How do the 98 and 86 versions differ, and are all differences covered?
Wubi 98 (Wangma 98) adjusted some character roots, so a minority of characters are encoded differently. This tool bundles only a small set of checkable differences (for example 母: XGUI in 86, XNNY in 98). In 98 mode those characters use the 98 code and every other character falls back to the 86 code, labelled "按 86 版". For a complete comparison consult the official Wangma table; nothing here is inferred.
How are spaces, punctuation and Latin letters handled in whole-text mode?
Spaces, newlines and tabs are ignored and are not counted. Punctuation, Latin letters and digits that are not in the table are reported as "not included"; compact mode only joins the same per-character results into one space-separated line, so its labels stay identical to one-line-per-character mode.
Why do different references give different codes for the same character?
Three common reasons: a different version (86 / 98 / New Century), mixing short codes with full codes, and writing key-name characters as a single key (人 as W) instead of four (人 WWWW). This tool always prints the 86 full code and states explicitly when it falls back in 98 mode.
Is my input uploaded anywhere?
No. The table ships with the page and conversion runs locally with pure functions: no network requests, no system clock, and nothing you type is sent or stored.
Keywords：五笔五笔编码五笔字型86版五笔98版五笔汉字转五笔五笔码查询wubiwubi codewubi 86wubi 98chinese input method codewubis