← Models

Open language infrastructure / Wikilangs

A world of language.

Open original playground ↗

One open collection. Hundreds of Wikipedia languages. Explore where they live, how they relate, and what their models can do.

360languages indexed
318with published models
92training groups
5model types

01 / The atlas

The atlas.

Choose a region to focus the catalog
Fiji · OceaniaTanzania · AfricaCanada · North AmericaUnited States of America · North AmericaKazakhstan · AsiaUzbekistan · AsiaPapua New Guinea · OceaniaIndonesia · AsiaArgentina · South AmericaChile · South AmericaDem. Rep. Congo · AfricaSomalia · AfricaKenya · AfricaSudan · AfricaChad · AfricaHaiti · North AmericaDominican Rep. · North AmericaRussia · EuropeBahamas · North AmericaFalkland Is. · South AmericaNorway · EuropeGreenland · North AmericaFr. S. Antarctic Lands · AfricaTimor-Leste · AsiaSouth Africa · AfricaLesotho · AfricaMexico · North AmericaUruguay · South AmericaBrazil · South AmericaBolivia · South AmericaPeru · South AmericaColombia · South AmericaPanama · North AmericaCosta Rica · North AmericaNicaragua · North AmericaHonduras · North AmericaEl Salvador · North AmericaGuatemala · North AmericaBelize · North AmericaVenezuela · South AmericaGuyana · South AmericaSuriname · South AmericaFrance · EuropeEcuador · South AmericaPuerto Rico · North AmericaJamaica · North AmericaCuba · North AmericaZimbabwe · AfricaBotswana · AfricaNamibia · AfricaSenegal · AfricaMali · AfricaMauritania · AfricaBenin · AfricaNiger · AfricaNigeria · AfricaCameroon · AfricaTogo · AfricaGhana · AfricaCôte d'Ivoire · AfricaGuinea · AfricaGuinea-Bissau · AfricaLiberia · AfricaSierra Leone · AfricaBurkina Faso · AfricaCentral African Rep. · AfricaCongo · AfricaGabon · AfricaEq. Guinea · AfricaZambia · AfricaMalawi · AfricaMozambique · AfricaeSwatini · AfricaAngola · AfricaBurundi · AfricaIsrael · AsiaLebanon · AsiaMadagascar · AfricaPalestine · AsiaGambia · AfricaTunisia · AfricaAlgeria · AfricaJordan · AsiaUnited Arab Emirates · AsiaQatar · AsiaKuwait · AsiaIraq · AsiaOman · AsiaVanuatu · OceaniaCambodia · AsiaThailand · AsiaLaos · AsiaMyanmar · AsiaVietnam · AsiaNorth Korea · AsiaSouth Korea · AsiaMongolia · AsiaIndia · AsiaBangladesh · AsiaBhutan · AsiaNepal · AsiaPakistan · AsiaAfghanistan · AsiaTajikistan · AsiaKyrgyzstan · AsiaTurkmenistan · AsiaIran · AsiaSyria · AsiaArmenia · AsiaSweden · EuropeBelarus · EuropeUkraine · EuropePoland · EuropeAustria · EuropeHungary · EuropeMoldova · EuropeRomania · EuropeLithuania · EuropeLatvia · EuropeEstonia · EuropeGermany · EuropeBulgaria · EuropeGreece · EuropeTurkey · AsiaAlbania · EuropeCroatia · EuropeSwitzerland · EuropeLuxembourg · EuropeBelgium · EuropeNetherlands · EuropePortugal · EuropeSpain · EuropeIreland · EuropeNew Caledonia · OceaniaSolomon Is. · OceaniaNew Zealand · OceaniaAustralia · OceaniaSri Lanka · AsiaChina · AsiaTaiwan · AsiaItaly · EuropeDenmark · EuropeUnited Kingdom · EuropeIceland · EuropeAzerbaijan · AsiaGeorgia · AsiaPhilippines · AsiaMalaysia · AsiaBrunei · AsiaSlovenia · EuropeFinland · EuropeSlovakia · EuropeCzechia · EuropeEritrea · AfricaJapan · AsiaParaguay · South AmericaYemen · AsiaSaudi Arabia · AsiaAntarcticaN. CyprusCyprus · AsiaEgypt · AfricaLibya · AfricaEthiopia · AfricaDjibouti · AfricaSomalilandUganda · AfricaRwanda · AfricaBosnia and Herz. · EuropeMacedonia · EuropeSerbia · EuropeMontenegro · EuropeKosovoTrinidad and Tobago · North AmericaS. Sudan · AfricaMorocco · Africa

Map colors show broad regional association. Some languages span regions. Counts refer to published models.

02 / Connections

Language families.

These groups combine linguistic relationships and script for multilingual model training. Select one to see its languages.

Training groups

92 groups

03 / Language explorer

Explore a language.

ary / Africa

Moroccan Arabic

الدارجة

Model card ↗
Training group
Vocabulary78,779 tokens
Best compression4.17×

From the Moroccan Arabic Wikipedia edition

01هادي صفحة د التوضيح، كلمة بركان يمكن يكونو عندها هاد لمعاني: بْرْكان: مدينة مغري...
02لْفزضاض ؤلا أفزضاض (سمية لعلمية Microcosmus sabatieri) حيوان لاسنسولي كيعيش ف لب...
03نيلز أبراهام لانݣليت (مزيود ف 9 يوليوز - مات ف 30 مارس هوّا عالم د شّيمي سويدي. ...

Nearby in Arabic

Read 19 recorded model continuations

Moroccan Arabic ary

د دّواور لعاداد د سّكان ديالو واصل ل 71 شخص على حساب لإحصاء الرسمي د عام نوطات مصادر ف لمغريب هاد دّوار كينتامي ل مشيخة ايت وادوز العزابة لي كتضم

Three-gram continuation · seed: د دّواور

Kabyle kab

tamaneɣt nnes d tamdint n wegmuḍ ccerq n tmurt n spanya tawurta

Three-gram continuation · seed: tamaneɣt nnes

Standard Moroccan Tamazight zgh

démographiques et socio économiques de la population rurale hors nomades par douar selon le recensement général de la population rurale hors nomades par douar selon le recensement général de la

Three-gram continuation · seed: démographiques et

Cherokee chr

ꭰᏸꮅ be checked ꭱꮃꮧꮬ ꮖꮗꭲꮴꭹꭲꮒ

Three-gram continuation · seed: ꭰᏸꮅ be

Quechua qu

distritu pruwinsya kastilla simita rimaqkuna 1 indihina simita rimaqkuna 1 2 challwawachu

Three-gram continuation · seed: distritu pruwinsya

Maori mi

he tāone tonu tēnei ko te takiwā o manawatū whanganui rāngai tūmatanui canterbury district health board 10 rongoā

Three-gram continuation · seed: he tāone

Hawaiian haw

mea paʻakikī kona hele aku i ka manawa o ka huinakolu polenekia

Three-gram continuation · seed: mea paʻakikī

Swahili sw

ya watakatifu wa agano la kale orodha ya mito ya mkoa wa dola la uturuki ambalo

Three-gram continuation · seed: ya watakatifu

Zulu zu

the national government constituencies development fund ng cdf kulandwe ngo zibandlela 9 imithombo zase bhurundi

Three-gram continuation · seed: the national

Yoruba yo

ìgbàjá ástẹ́rọ́ìdì itokasi ástẹ́rọ́ìdì vec lista de países por pib

Three-gram continuation · seed: ìgbàjá ástẹ́rọ́ìdì

Tibetan bo

པ ཞེས འབོད སྲོལ ཡོད པ དེ ལ གོ བ ལྟར བོད དུ ཕེབས བསུ བ རྒྱ གར གྱི མཁས པ གཅིག ཀྱང མི ཤེས པ ཡིན པ དང རྡུལ

Three-gram continuation · seed: པ ཞེས

Irish ga

ann breatnais a labhairt na breataine é an 20ú haois cé nach bhfuil ach iseatóip radaighníomhacha aige

Three-gram continuation · seed: ann breatnais

Finnish fi

lemmon mt lemmon survey 11 maaliskuuta kitt peak spacewatch 19 lokakuuta mount lemmon survey 20 lokakuuta 10 lokakuuta mount lemmon mt lemmon survey 23 helmikuuta catalina css 13 joulukuuta helsinki

Three-gram continuation · seed: lemmon mt

Tamil ta

ஆண்டுத் தரவின்படி கீழ்க்கண்ட தகவல் தொகுக்கப்பட்டுள்ளது அடிப்படை வசதிகள் தமிழ்நாடு ஊரக வளர்ச்சி மற்றும் ஊராட்சித் துறையின் ஆண்டுத் தரவின்படி கீழ்க்கண்ட தகவல் தொகுக்கப்பட்டுள்ளது அடிப்படை வசதிகள் எண்ணிக்கை குடிநீர் இணைப்புகள் 317 சிறு மின்விசைக் குழாய்கள் 10 கைக்குழாய்கள் 22

Three-gram continuation · seed: ஆண்டுத் தரவின்படி

Amharic am

በዘመነ ማርቆስና በዘመነ ሉቃስ ዕለታት ሲቀሩ በዘመነ ዮሐንስ ዘመነ ማቴዎስ እና ዘመነ ማርቆስ ደግሞ ቀናት ይቀራሉ ዓቢይ ታሪካዊ ማስታወሻዎች ዓ ም በኋላ ለሆኑት ዓመታት ግን በሌላ ቀን ላይ መሆኑን ይገንዘቡ

Three-gram continuation · seed: በዘመነ ማርቆስና

Welsh cy

ffilmiau gan gynnwys y canlynol ffilm delwedd gwlad iaith wreiddiol dyddiad fuga nella tempesta yr eidal eidaleg cyfeiriadau gan gyfarwyddwyr ffilm gwrywaidd saesneg lliw lliw o unol daleithiau america sgwennwyd

Three-gram continuation · seed: ffilmiau gan

Bengali bn

ধারণা করা হয় এবং এর দশকের শেষদিকে ইংল্যান্ডের বিপক্ষে মাত্র তিন বছর বয়সী এবং তার চাচাতো ভাই শেখ হেলাল উদ্দীন আব্বাস

Three-gram continuation · seed: ধারণা করা

Indonesian id

hidup atau kayu yang telah rusak dan tidak mencari tahu bagaimana cara menggunakan aplikasi tersebut juga dapat menyerap oksigen dari air laut sampai ke sana karier politik jmpl lurus ka

Three-gram continuation · seed: hidup atau

Javanese jv

jawa tengah réferènsi ing nurwègen réferènsi ing kabupatèn bengkalis riau indonésia désa iki kapérang dadi golongan iki duwé peran sajeroning

Three-gram continuation · seed: jawa tengah

Recorded sequences from published Wikilangs three-gram models. These are word transitions, not assistant responses.

Catalog, metrics, and excerpts: Wikilangs published language metadata. Model training groups: BabelVec family assignments, which combine language relationships and script. The live playground runs at wikilangs.org. Explore the source project ↗