Complete Gujarati varnamala (kakko) chart with swar, vyanjan, signs, barakhadi, conjuncts and numerals, each Gujarati letter shown beside its Devanagari equivalent (the script Hindi uses). Tap any letter to hear it, and print the chart.
Your inputs stay in your browser - nothing you type is sent to a server
The clearest answer is a derived breakdown: 12 swar (vowels) plus 34 vyanjan (consonants) = 46. That is the modern orthographic inventory (W3C plus the Unicode Gujarati block), shown as a breakdown rather than a figure any authority decrees, since Gujarati has no akademi that enumerates an official count. The vocalic ઋ is a current Sanskritic vowel but is listed separately from the 12 plain vowels. You may also see 47 (a traditional convention that counts ઋ and the two signs among the swar) or 48 (some primer charts): both are common conventions rather than sourced standards, explained below.
Gujarati vs Devanagari notes: Gujarati and Devanagari share the same varga order, so each Gujarati letter maps almost 1:1 to a Devanagari slot (ક to क, ળ to ळ). The column is labelled Devanagari, not Hindi, because a few slots such as ળ to ळ are valid Devanagari letters but not standard modern-Hindi ones. The biggest visual difference is that Gujarati has no shirorekha, the top headline stroke that joins Devanagari letters. The 34 basic consonants are the 33 classical letters plus ળ (retroflex la). The conjuncts ક્ષ and જ્ઞ are shown below as jodakshar, not counted among the 34.
These two are diacritic marks that modify a sound and are not pronounced on their own. Unicode and W3C class them as combining marks, but some teaching charts count them in the swar row, which is how the 13-swar total is reached. Tap a sign to read what it does.
અનુસ્વાર (anusvara)
Tap to learn more
વિસર્ગ (visarga)
Tap to learn more
Barakhadi is the table children learn after the kakko: each consonant run through the traditional 12-form row (for example ક gives ક કા કિ કી કુ કૂ કે કૈ કો કૌ કં કઃ). Those 12 forms are the 10 core vowels plus anusvara and visarga; the row omits the vocalic ઋ and the two candra vowels. It is a common pedagogical sequence (sources vary on how many rows are drilled). Here are three sample rows; every consonant follows the same 12-form pattern.
Jodakshar are single glyphs formed by joining two consonants (for example ક + ્ + ષ gives ક્ષ). They are conjunct sequences, not counted among the 34 basic consonants. Tap to hear each one.
Gujarati and Devanagari both descend from the Brahmi script and share the same vowel-then-varga order, so the Devanagari column above maps each Gujarati letter to its slot almost one to one: અ to अ, ક to क, પ to प, and the extra Gujarati consonant ળ to ळ. Most of these slots are also standard Hindi letters, which is why the column is useful to Hindi readers, but a few (like ળ to ळ) are Devanagari letters that modern Hindi does not use, so the column is labelled Devanagari rather than Hindi. The clearest visible difference is that Gujarati drops the shirorekha, the horizontal headline stroke that runs across the top of Devanagari letters and binds them into words. The two candra vowels ઍ and ઑ map to the Devanagari candra vowels ऍ and ऑ, used for the same English-loan sounds. Turn the Devanagari column on or off with the button above.
Sequence follows the Gujarat GCERT / SCERT Standard-1 Gujarati primer (kakko) tradition. The board sets the teaching order; the letter counts above come from the modern orthographic inventory (W3C plus the Unicode Gujarati block) cross-referenced with Wikipedia, not from an official government count.
Hindi varnamala chart with all 46 letters per Govt of India: swar, vyanjan, sanyukt akshar, special signs and numerals, with tap-to-hear audio.
hindi varnamalaTelugu varnamala chart with achulu, hallulu, guninthalu, yogavahalu and numerals. Tap any letter to hear it, with a print-friendly chart.
telugu varnamalaBengali varnamala chart with swarabarna, byanjanbarna, signs, yuktakshar and numerals, each letter shown beside its Hindi (Devanagari) equivalent. Tap to hear or print.
bengali varnamalaTL;DR
The clearest answer to "how many letters in the Gujarati varnamala" is a derived breakdown: 12 swar (vowels) + 34 vyanjan (consonants) = 46. That is the modern orthographic inventory from W3C and the Unicode Gujarati block, shown as a breakdown rather than a figure any authority declares, because Gujarati has no akademi that enumerates an official count. The vocalic ઋ is a current Sanskritic vowel but is listed separately from the 12 plain vowels; a traditional count reaches 47 by counting ઋ and the two signs (anusvara, visarga) among the swar, and some primer charts show 48. Both 47 and 48 are common conventions rather than sourced standards (Unicode and W3C class those two signs as diacritic marks, not vowels). This page lists every letter beside its Devanagari equivalent (the script Hindi uses), with tap-to-hear pronunciation through your browser's Gujarati voice (where available), a barakhadi sample, jodakshar, and a one-tap browser print for classroom or home use.
The different counts come from what is being counted, not from any error. The clearest modern figure is a derived 46: 12 plain independent vowels plus 34 basic consonants, from the W3C modern orthography and the Unicode Gujarati block. The 12 vowels are the 10 core vowels plus the 2 candra vowels (ઍ and ઑ, used mainly for English-loan sounds). The vocalic ઋ is still in current use (for example ઋતુ, season), but W3C lists it apart from the 12 plain vowels, so it is shown separately rather than folded into the 12. Some teaching charts reach 47 by counting ઋ and the two signs anusvara (અં) and visarga (અઃ) in the swar row, for 13 swar plus 34 consonants. Unicode and W3C classify those two signs as combining marks rather than vowels, so 47 is a chart convention rather than a sourced enumeration. Some primer charts instead show 48 (a 14-swar chart convention whose two extra members vary from chart to chart); no single authoritative 14-member vowel set was identified, so this page pins neither 47 nor 48 as a count. The consonant total stays 34 in every case, and none of these is a government standard: Gujarati has no akademi that decrees an official letter count.
Gujarati and Devanagari both descend from the Brahmi script, so they share the same vowel-then-varga structure and map almost one to one: અ maps to अ, ક to क, પ to प, and Gujarati's extra consonant ળ to ळ, across the whole base set. Most of these Devanagari slots are also standard Hindi letters, which is why the column helps Hindi readers, but a few (like ળ to ळ, and the candra vowels below) are Devanagari letters modern Hindi does not use, so the column is labelled Devanagari rather than Hindi. Two differences are worth knowing. First, Gujarati has no shirorekha: Devanagari joins its letters with a horizontal headline stroke along the top, and Gujarati simply drops it, which is the fastest way to tell the two scripts apart at a glance. Second, the two candra vowels ઍ and ઑ (which map to Devanagari ऍ and ऑ) carry the English-loan "a" and "o" sounds, as in words borrowed from English. Use the Show / Hide Devanagari column button to turn this side-by-side view on or off.
Like other Brahmi-derived scripts, Gujarati orders its consonants by where each sound is produced in the mouth, moving from the throat to the lips. The 25 core consonants form five vargas:
After the vargas come the antastha and ushma letters, then ળ. The conjuncts ક્ષ (ksha) and જ્ઞ (gna) are jodakshar, formed by joining two consonants, and are not counted among the 34 basic consonants even though many charts list them at the end. Once the kakko is known, children move on to the barakhadi (બારાખડી), the table that runs each consonant through the traditional 12-form row (the 10 core vowels plus anusvara and visarga; it omits the vocalic ઋ and the two candra vowels).
Audio playback uses your browser's built-in Gujarati voice (Web Speech API) when one is installed; it is not a native recording, and pronunciation quality depends on your device. There is deliberately no non-Gujarati voice fallback, because a non-Gujarati voice would mispronounce these letters. No audio files are hosted on this site. The sourced 46 count shown above is derived from the tool's data file and pinned by unit tests in __tests__/lib/gujarati-varnamala-data.test.ts; the 47 and 48 conventions are described in prose, not pinned. Last reviewed and sources accessed on July 18, 2026.