[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"footer-settings":3,"main-navigation":153,"footer-homepage":836,"footer-integrations":903,"blog-post-nav-nav":1814,"blog-post-nav-[USE LATEST FEATURED]":1836,"blog-ai-8-llm-arabic-models-tested-to-translate":1851},{"id":4,"status":5,"security_section_title":6,"loved_section_title":7,"columns":8,"social_links":113,"policy_links":143},1,"published","Secured & trusted","Loved by users",[9,28,48,74,78],{"id":4,"status":5,"sort":10,"title":11,"column_key":12,"links":13},null,"Localazy","connect",[14,18,23],{"id":15,"status":5,"sort":4,"title":16,"url":17},11,"Book a demo","___LOCALAZY_MEETING_URL___",{"id":19,"status":5,"sort":20,"title":21,"url":22},12,2,"Contact","\u002Fcontact",{"id":24,"status":5,"sort":25,"title":26,"url":27},21,3,"About us","\u002Fabout-us",{"id":20,"status":5,"sort":10,"title":29,"column_key":30,"links":31},"Programs","programs",[32,36,40,44],{"id":4,"status":5,"sort":33,"title":34,"url":35},4,"Partner Program","\u002Fpartnership",{"id":20,"status":5,"sort":37,"title":38,"url":39},5,"Ambassador Program","\u002Fambassador",{"id":25,"status":5,"sort":41,"title":42,"url":43},6,"Startup Program","\u002Ffor\u002Fstartups",{"id":33,"status":5,"sort":45,"title":46,"url":47},7,"Nonprofits","\u002Ffor\u002Fnonprofit",{"id":25,"status":5,"sort":10,"title":49,"column_key":50,"links":51},"Use Cases","use_cases",[52,56,60,64,67,70],{"id":37,"status":5,"sort":53,"title":54,"url":55},8,"Software Localization","\u002Fterm\u002Fsoftware-localization",{"id":41,"status":5,"sort":57,"title":58,"url":59},9,"Machine Translation Services","\u002Ffeatures\u002Fmachine-translation",{"id":45,"status":5,"sort":61,"title":62,"url":63},10,"Translation API","\u002Ffeatures\u002Ftranslation-api",{"id":53,"status":5,"sort":15,"title":65,"url":66},"Crowdsourced Translations","\u002Ffeatures\u002Fshare-tm",{"id":57,"status":5,"sort":19,"title":68,"url":69},"Figma i18n & localization","\u002Ffeatures\u002Ffigma-localization-plugin",{"id":61,"status":5,"sort":71,"title":72,"url":73},13,"Translation as a Service","\u002Fterm\u002Ftranslation-service",{"id":33,"status":5,"sort":10,"title":75,"column_key":76,"links":77},"Integrations","integrations",[],{"id":37,"status":5,"sort":10,"title":79,"column_key":80,"links":81},"Resources","resources",[82,86,90,94,98,102,106,110],{"id":71,"status":5,"sort":83,"title":84,"url":85},14,"Documentation","\u002Fdocs",{"id":83,"status":5,"sort":87,"title":88,"url":89},15,"Dictionary","\u002Fdictionary",{"id":87,"status":5,"sort":91,"title":92,"url":93},16,"Case Studies","\u002Fcase-study",{"id":91,"status":5,"sort":95,"title":96,"url":97},17,"Discussion forum","___DISCUSS_URL___",{"id":95,"status":5,"sort":99,"title":100,"url":101},18,"Localization Blog","\u002Fblog",{"id":99,"status":5,"sort":103,"title":104,"url":105},19,"FAQ","\u002Ffaq",{"id":103,"status":5,"sort":107,"title":108,"url":109},20,"Pricing","\u002Fpricing",{"id":107,"status":5,"sort":24,"title":111,"url":112},"Brand assets","https:\u002F\u002Fdrive.google.com\u002Fdrive\u002Fu\u002F1\u002Ffolders\u002F1tZqsZHBlGelz7A2FwTsiwrs4j3sd9HTk",[114,119,124,129,134,139],{"id":4,"status":5,"sort":10,"platform":115,"icon":116,"url":117,"label":118},"feed","rss","\u002Ffeed.xml","Localazy RSS Feed",{"id":20,"status":5,"sort":10,"platform":120,"icon":121,"url":122,"label":123},"facebook","fb-logo","https:\u002F\u002Fwww.facebook.com\u002Flocalazy\u002F","Localazy facebook",{"id":25,"status":5,"sort":10,"platform":125,"icon":126,"url":127,"label":128},"github","github-logo","https:\u002F\u002Fgithub.com\u002Flocalazy","Localazy GitHub",{"id":33,"status":5,"sort":10,"platform":130,"icon":131,"url":132,"label":133},"x","x-twitter","https:\u002F\u002Fx.com\u002Flocalazy","Localazy X",{"id":37,"status":5,"sort":10,"platform":135,"icon":136,"url":137,"label":138},"linkedin","linkedin-logo","https:\u002F\u002Fwww.linkedin.com\u002Fcompany\u002F37836599\u002F","Localazy LinkedIn",{"id":41,"status":5,"sort":10,"platform":140,"icon":141,"url":97,"label":142},"discourse","discourse-logo","Localazy Discourse",[144,147,150],{"id":4,"status":5,"sort":10,"title":145,"url":146,"action":10},"Privacy policy","\u002Fprivacy-policy",{"id":20,"status":5,"sort":10,"title":148,"url":149,"action":10},"Terms & Conditions","\u002Fterms-and-conditions",{"id":25,"status":5,"sort":10,"title":151,"url":10,"action":152},"Cookies","cookie-consent",[154,347,580,708,831,833],{"id":20,"name":155,"url":10,"url_text":10,"status":5,"groups":156},"Solutions",[157,168,239,274],{"navigation_items_group_id":158},{"id":25,"sort":37,"status":5,"name":159,"render_show_all_url":160,"show_all_url_position":161,"show_favorite_icon":162,"show_all_url":10,"represent_as_tile":160,"items":163},"Featured Article",false,"top",true,[164],{"collection":165,"item":166},"blog_post_navigation_item_type",{"id":33,"status":5,"sort":10,"show_author_img":162,"blog_post_tag":167,"description":10},"nav",{"navigation_items_group_id":169},{"id":37,"sort":4,"status":5,"name":170,"render_show_all_url":160,"show_all_url_position":161,"show_favorite_icon":160,"show_all_url":171,"represent_as_tile":162,"items":172},"Your Role","\u002Froles",[173,185,195,205,216,228],{"collection":174,"item":175},"relation_navigation_item_type",{"id":4,"status":5,"sort":10,"name":176,"item":177},"CMO",[178],{"collection":179,"item":180},"personas",{"id":4,"status":5,"sort":10,"slug":181,"icon":182},"chief-marketing-officers",{"id":183,"title":184},"526097d8-176f-48e5-83d3-2fc4d77061f9","La Mail Bulk",{"collection":174,"item":186},{"id":25,"status":5,"sort":10,"name":187,"item":188},"CTO",[189],{"collection":179,"item":190},{"id":33,"status":5,"sort":10,"slug":191,"icon":192},"chief-technical-officers",{"id":193,"title":194},"f808fdf0-c78b-4294-ba59-fe25f46fe825","La Mechanic",{"collection":174,"item":196},{"id":107,"status":5,"sort":10,"name":197,"item":198},"Product Manager",[199],{"collection":179,"item":200},{"id":99,"status":5,"sort":10,"slug":201,"icon":202},"product-managers",{"id":203,"title":204},"ea39586a-21fc-4b34-ba87-496028e5198f","La Worker",{"collection":174,"item":206},{"id":207,"status":5,"sort":10,"name":208,"item":209},37,"Translator",[210],{"collection":179,"item":211},{"id":107,"status":5,"sort":10,"slug":212,"icon":213},"professional-translators",{"id":214,"title":215},"07945171-d51d-4ca8-96c0-21cde9db258f","La Chat",{"collection":174,"item":217},{"id":218,"status":5,"sort":10,"name":219,"item":220},38,"Developer",[221],{"collection":179,"item":222},{"id":223,"status":5,"sort":10,"slug":224,"icon":225},30,"software-developers",{"id":226,"title":227},"1806e46d-c551-4845-bf42-4407344d229e","La Terminal",{"collection":174,"item":229},{"id":230,"status":5,"sort":10,"name":231,"item":232},27,"Localization Manager",[233],{"collection":179,"item":234},{"id":41,"status":5,"sort":10,"slug":235,"icon":236},"localization-managers",{"id":237,"title":238},"dac69bd1-bf5d-41d0-87d4-dd13e314698a","La Language",{"navigation_items_group_id":240},{"id":57,"sort":25,"status":5,"name":241,"render_show_all_url":160,"show_all_url_position":242,"show_favorite_icon":160,"show_all_url":10,"represent_as_tile":160,"items":243},"Use cases","bottom",[244,250,255,260,264,269],{"collection":245,"item":246},"simple_link_navigation_item_type",{"id":247,"status":5,"sort":10,"url":248,"url_text":249,"show_favorite_icon":160},24,"https:\u002F\u002Flocalazy.com\u002Fuse-cases\u002Fsoftware-localization","Software localization",{"collection":245,"item":251},{"id":252,"status":5,"sort":10,"url":253,"url_text":254,"show_favorite_icon":160},26,"https:\u002F\u002Flocalazy.com\u002Fuse-cases\u002Fmobile-app-localization","Mobile app localization",{"collection":245,"item":256},{"id":257,"status":5,"sort":10,"url":258,"url_text":259,"show_favorite_icon":160},25,"https:\u002F\u002Flocalazy.com\u002Fuse-cases\u002Fwebsite-localization","Website localization",{"collection":245,"item":261},{"id":230,"status":5,"sort":10,"url":262,"url_text":263,"show_favorite_icon":160},"https:\u002F\u002Flocalazy.com\u002Fuse-cases\u002Fdesign-localization","Design localization",{"collection":245,"item":265},{"id":266,"status":5,"sort":10,"url":267,"url_text":268,"show_favorite_icon":160},28,"https:\u002F\u002Flocalazy.com\u002Fuse-cases\u002Fgame-localization","Game localization",{"collection":245,"item":270},{"id":271,"status":5,"sort":10,"url":272,"url_text":273,"show_favorite_icon":160},29,"https:\u002F\u002Flocalazy.com\u002Fuse-cases\u002Fhelp-and-docs-localization","Help & docs localization",{"navigation_items_group_id":275},{"id":53,"sort":20,"status":5,"name":276,"render_show_all_url":160,"show_all_url_position":242,"show_favorite_icon":160,"show_all_url":10,"represent_as_tile":160,"items":277},"Industry",[278,289,300,310,320,329,340],{"collection":174,"item":279},{"id":280,"status":5,"sort":10,"name":281,"item":282},23,"Digital Agencies",[283],{"collection":179,"item":284},{"id":45,"status":5,"sort":10,"slug":285,"icon":286},"digital-agencies",{"id":287,"title":288},"547f5619-b21d-4b29-86f1-8b2200313cde","La Digi Agency",{"collection":174,"item":290},{"id":291,"status":5,"sort":10,"name":292,"item":293},39,"Startups & SaaS",[294],{"collection":179,"item":295},{"id":20,"status":5,"sort":10,"slug":296,"icon":297},"startups",{"id":298,"title":299},"da1516e2-4afc-40a2-bcbf-a8161e2d610d","La Saa S",{"collection":174,"item":301},{"id":20,"status":5,"sort":10,"name":302,"item":303},"E-commerce",[304],{"collection":179,"item":305},{"id":37,"status":5,"sort":10,"slug":306,"icon":307},"e-commerce",{"id":308,"title":309},"a3d09cd2-054f-4bb5-a0a2-d3d207db65df","La Ecommerce",{"collection":174,"item":311},{"id":103,"status":5,"sort":10,"name":312,"item":313},"FinTech Platforms",[314],{"collection":179,"item":315},{"id":230,"status":5,"sort":10,"slug":316,"icon":317},"fintech",{"id":318,"title":319},"147339ba-07d1-4e5e-baad-23d27e60c919","La Money",{"collection":174,"item":321},{"id":247,"status":5,"sort":10,"name":322,"item":323},"Translation Agencies",[324],{"collection":179,"item":325},{"id":247,"status":5,"sort":10,"slug":326,"icon":327},"translation-agencies",{"id":328,"title":238},"15d6fe33-2af8-4d6e-933d-5caa64394511",{"collection":174,"item":330},{"id":331,"status":5,"sort":10,"name":332,"item":333},22,"Travel & Tourism",[334],{"collection":179,"item":335},{"id":25,"status":5,"sort":10,"slug":336,"icon":337},"tourism",{"id":338,"title":339},"23a785e8-0ae5-4c4b-8452-aab1b58182a6","La Travel",{"collection":174,"item":341},{"id":257,"status":5,"sort":10,"name":342,"item":343},"Nonprofit Causes",[344],{"collection":179,"item":345},{"id":266,"status":5,"sort":10,"slug":346,"icon":10},"nonprofit",{"id":37,"name":348,"url":10,"url_text":10,"status":5,"groups":349},"Features",[350,417,474,523],{"navigation_items_group_id":351},{"id":15,"sort":41,"status":5,"name":352,"render_show_all_url":160,"show_all_url_position":161,"show_favorite_icon":160,"show_all_url":10,"represent_as_tile":160,"items":353},"Localization",[354,363,372,380,388,392,400,409],{"collection":174,"item":355},{"id":356,"status":5,"sort":10,"name":357,"item":358},44,"Localization services",[359],{"collection":360,"item":361},"products",{"id":57,"status":5,"slug":362},"continuous-localization-team",{"collection":174,"item":364},{"id":365,"status":5,"sort":10,"name":366,"item":367},45,"Translation management",[368],{"collection":369,"item":370},"term_landing_page",{"id":271,"status":5,"slug":371},"translation-management-system",{"collection":174,"item":373},{"id":374,"status":5,"sort":10,"name":375,"item":376},46,"Context Screenshots",[377],{"collection":360,"item":378},{"id":71,"status":5,"slug":379},"context-screenshots-ocr",{"collection":174,"item":381},{"id":382,"status":5,"sort":10,"name":383,"item":384},47,"Translation Glossary",[385],{"collection":360,"item":386},{"id":95,"status":5,"slug":387},"glossary",{"collection":245,"item":389},{"id":33,"status":5,"sort":10,"url":390,"url_text":391,"show_favorite_icon":160},"\u002Fdocs\u002Fgeneral\u002Freviewing-translations","Quality control",{"collection":174,"item":393},{"id":394,"status":5,"sort":10,"name":395,"item":396},34,"Connected Projects",[397],{"collection":360,"item":398},{"id":4,"status":5,"slug":399},"connected-projects",{"collection":174,"item":401},{"id":402,"status":5,"sort":10,"name":403,"item":404},64,"Plural handling",[405],{"collection":360,"item":406},{"id":407,"status":5,"slug":408},32,"plurals",{"collection":174,"item":410},{"id":411,"status":5,"sort":10,"name":412,"item":413},65,"Style guides",[414],{"collection":360,"item":415},{"id":218,"status":5,"slug":416},"style-guide",{"navigation_items_group_id":418},{"id":19,"sort":45,"status":5,"name":419,"render_show_all_url":160,"show_all_url_position":161,"show_favorite_icon":160,"show_all_url":10,"represent_as_tile":160,"items":420},"Translation",[421,429,437,441,449,457,466],{"collection":174,"item":422},{"id":423,"status":5,"sort":10,"name":424,"item":425},36,"Professional translations",[426],{"collection":369,"item":427},{"id":223,"status":5,"slug":428},"professional-translation-services",{"collection":174,"item":430},{"id":431,"status":5,"sort":10,"name":432,"item":433},48,"Crowdsourced translations",[434],{"collection":360,"item":435},{"id":53,"status":5,"slug":436},"share-tm",{"collection":245,"item":438},{"id":37,"status":5,"sort":10,"url":439,"url_text":440,"show_favorite_icon":160},"\u002Fdocs\u002Fgeneral\u002Ftranslating-strings","Translation interface",{"collection":174,"item":442},{"id":443,"status":5,"sort":10,"name":444,"item":445},55,"Machine translations",[446],{"collection":360,"item":447},{"id":37,"status":5,"slug":448},"machine-translation",{"collection":174,"item":450},{"id":451,"status":5,"sort":10,"name":452,"item":453},60,"Localazy AI translation",[454],{"collection":360,"item":455},{"id":271,"status":5,"slug":456},"localazy-ai",{"collection":174,"item":458},{"id":459,"status":5,"sort":10,"name":460,"item":461},61,"Translation Memory",[462],{"collection":360,"item":463},{"id":464,"status":5,"slug":465},35,"translation-memory",{"collection":174,"item":467},{"id":468,"status":5,"sort":10,"name":469,"item":470},63,"Code & placeholders",[471],{"collection":360,"item":472},{"id":223,"status":5,"slug":473},"code-and-placeholders",{"navigation_items_group_id":475},{"id":71,"sort":53,"status":5,"name":476,"render_show_all_url":160,"show_all_url_position":161,"show_favorite_icon":160,"show_all_url":10,"represent_as_tile":160,"items":477},"Management",[478,482,490,498,507,515,519],{"collection":245,"item":479},{"id":53,"status":5,"sort":10,"url":480,"url_text":481,"show_favorite_icon":160},"\u002Fdocs\u002Fgeneral\u002Fdefining-user-roles","User roles",{"collection":174,"item":483},{"id":484,"status":5,"sort":10,"name":485,"item":486},51,"Duplicity Linking",[487],{"collection":360,"item":488},{"id":103,"status":5,"slug":489},"duplicity-linking",{"collection":174,"item":491},{"id":492,"status":5,"sort":10,"name":493,"item":494},52,"Language Permissions",[495],{"collection":360,"item":496},{"id":24,"status":5,"slug":497},"language-permissions",{"collection":174,"item":499},{"id":500,"status":501,"sort":10,"name":502,"item":503},53,"draft","Automations",[504],{"collection":360,"item":505},{"id":107,"status":5,"slug":506},"automations",{"collection":174,"item":508},{"id":509,"status":5,"sort":10,"name":510,"item":511},58,"Branching",[512],{"collection":360,"item":513},{"id":230,"status":5,"slug":514},"branching",{"collection":245,"item":516},{"id":24,"status":501,"sort":10,"url":517,"url_text":518,"show_favorite_icon":160},"https:\u002F\u002Flocalazy.com\u002Fdocs\u002Fgeneral\u002Fsso-configuration","SSO",{"collection":245,"item":520},{"id":331,"status":501,"sort":10,"url":521,"url_text":522,"show_favorite_icon":160},"https:\u002F\u002Flocalazy.com\u002Fdocs\u002Fgeneral\u002Flanguage-statistics","Reporting",{"navigation_items_group_id":524},{"id":83,"sort":57,"status":5,"name":525,"render_show_all_url":160,"show_all_url_position":161,"show_favorite_icon":160,"show_all_url":10,"represent_as_tile":160,"items":526},"Development",[527,533,541,548,555,563,572],{"collection":174,"item":528},{"id":271,"status":5,"sort":10,"name":62,"item":529},[530],{"collection":360,"item":531},{"id":15,"status":5,"slug":532},"translation-api",{"collection":174,"item":534},{"id":535,"status":5,"sort":10,"name":536,"item":537},33,"Localazy CLI",[538],{"collection":360,"item":539},{"id":45,"status":5,"slug":540},"cli",{"collection":174,"item":542},{"id":407,"status":5,"sort":10,"name":543,"item":544},"Language CDN",[545],{"collection":360,"item":546},{"id":33,"status":5,"slug":547},"language-cdn",{"collection":174,"item":549},{"id":266,"status":5,"sort":10,"name":550,"item":551},"Releases",[552],{"collection":360,"item":553},{"id":41,"status":5,"slug":554},"releases",{"collection":174,"item":556},{"id":557,"status":501,"sort":10,"name":558,"item":559},57,"Format Conversions",[560],{"collection":360,"item":561},{"id":20,"status":5,"slug":562},"format-conversions",{"collection":174,"item":564},{"id":565,"status":5,"sort":10,"name":566,"item":567},59,"Webhooks",[568],{"collection":360,"item":569},{"id":570,"status":5,"slug":571},31,"webhooks",{"collection":174,"item":573},{"id":574,"status":5,"sort":10,"name":575,"item":576},62,"Export Aliases",[577],{"collection":360,"item":578},{"id":535,"status":5,"slug":579},"export-aliases",{"id":33,"name":75,"url":10,"url_text":10,"status":5,"groups":581},[582,698],{"navigation_items_group_id":583},{"id":45,"sort":61,"status":5,"name":584,"render_show_all_url":160,"show_all_url_position":161,"show_favorite_icon":160,"show_all_url":10,"represent_as_tile":160,"items":585},"Most popular",[586],{"collection":174,"item":587},{"id":19,"status":5,"sort":10,"name":75,"item":588},[589,597,604,611,618,624,631,638,645,652,659,668,675,682,690],{"collection":590,"item":591},"platforms",{"id":41,"status":5,"sort":4,"slug":592,"name":593,"icon":594},"android","Android",{"id":595,"title":596},"ff965add-53b0-44ad-b27c-ea4ef641d384","Ic Outline Android",{"collection":590,"item":598},{"id":45,"status":5,"sort":20,"slug":599,"name":600,"icon":601},"ios","iOS",{"id":602,"title":603},"886285d7-816a-4430-b5a5-edfe307b238e","Ios",{"collection":590,"item":605},{"id":535,"status":5,"sort":25,"slug":606,"name":607,"icon":608},"react","React",{"id":609,"title":610},"26757a90-88a3-4a94-b587-9f38614ed3bd","Logos React",{"collection":590,"item":612},{"id":103,"status":5,"sort":33,"slug":613,"name":614,"icon":615},"vuejs","Vue.js",{"id":616,"title":617},"6180b41f-3249-47d3-adcd-50fe1cf11bf1","Vue",{"collection":590,"item":619},{"id":91,"status":5,"sort":37,"slug":620,"name":621,"icon":622},"angular","Angular",{"id":623,"title":621},"2a31a97a-48c5-4531-aa36-41c9dd0bdbef",{"collection":590,"item":625},{"id":4,"status":5,"sort":41,"slug":626,"name":627,"icon":628},"flutter","Flutter",{"id":629,"title":630},"b9f13a11-9328-4a03-86cf-b20de0685606","Logos Flutter",{"collection":590,"item":632},{"id":25,"status":5,"sort":15,"slug":633,"name":634,"icon":635},"typescript","TypeScript",{"id":636,"title":637},"8d1319b5-8a96-4199-8407-00add3418b1f","Logos Typescript Icon",{"collection":590,"item":639},{"id":570,"status":5,"sort":19,"slug":640,"name":641,"icon":642},"javascript","JavaScript",{"id":643,"title":644},"4218d0c9-7e6b-494d-9663-37ceaf93ee8c","Logos Javascript",{"collection":590,"item":646},{"id":19,"status":5,"sort":107,"slug":647,"name":648,"icon":649},"json","JSON",{"id":650,"title":651},"99ac3b7c-dba5-4693-a4f4-1f27a6d7782d","Logos Json",{"collection":590,"item":653},{"id":24,"status":5,"sort":24,"slug":654,"name":655,"icon":656},"xliff","XLIFF",{"id":657,"title":658},"ca424cc8-e8a7-4aef-8dca-ef4a4d1334fc","Format",{"collection":590,"item":660},{"id":661,"status":5,"sort":662,"slug":663,"name":664,"icon":665},72,54,"strapi","Strapi",{"id":666,"title":667},"cde7f1d0-f2e0-4243-929e-a3f8f9146e69","Strapi Logo Purple",{"collection":590,"item":669},{"id":451,"status":5,"sort":443,"slug":670,"name":671,"icon":672},"figma","Figma",{"id":673,"title":674},"813745a4-a77d-4735-93b7-e0dd731b2304","Figma Icon",{"collection":590,"item":676},{"id":484,"status":5,"sort":382,"slug":677,"name":678,"icon":679},"microsoft-excel","Microsoft Excel",{"id":680,"title":681},"7a175955-7b96-4561-939f-a0e1924faa40","Excel",{"collection":590,"item":683},{"id":684,"status":5,"sort":459,"slug":685,"name":686,"icon":687},66,"github-actions","GitHub Actions",{"id":688,"title":689},"da17554f-ac6d-443d-9bac-1f29645ec1c6","Github Icon",{"collection":590,"item":691},{"id":692,"status":5,"sort":693,"slug":694,"name":695,"icon":696},97,999,"webflow","Webflow",{"id":697,"title":695},"ff43a386-52f8-499f-b77c-ab9a30f4a77e",{"navigation_items_group_id":699},{"id":33,"sort":15,"status":5,"name":700,"render_show_all_url":160,"show_all_url_position":161,"show_favorite_icon":160,"show_all_url":10,"represent_as_tile":160,"items":701},"Explore more",[702],{"collection":703,"item":704},"text_link_navigation_item_type",{"id":4,"status":5,"sort":10,"text":705,"url":706,"url_text":707},"Localazy is a technology-agnostic platform supporting 50+ frameworks, file formats & popular tools.","\u002Fintegrations","See all integrations",{"id":4,"name":79,"url":10,"url_text":10,"status":5,"groups":709},[710,734,740],{"navigation_items_group_id":711},{"id":4,"sort":19,"status":5,"name":712,"render_show_all_url":160,"show_all_url_position":161,"show_favorite_icon":160,"show_all_url":10,"represent_as_tile":160,"items":713},"Helpful Links",[714,717,719,723,725,728,730],{"collection":245,"item":715},{"id":71,"status":5,"sort":10,"url":101,"url_text":716,"show_favorite_icon":162},"Localazy Blog",{"collection":245,"item":718},{"id":83,"status":5,"sort":10,"url":85,"url_text":84,"show_favorite_icon":162},{"collection":245,"item":720},{"id":4,"status":5,"sort":10,"url":721,"url_text":722,"show_favorite_icon":160},"https:\u002F\u002Fdiscuss.localazy.com\u002F","Discussion Forum",{"collection":245,"item":724},{"id":19,"status":5,"sort":10,"url":93,"url_text":92,"show_favorite_icon":160},{"collection":245,"item":726},{"id":25,"status":5,"sort":10,"url":35,"url_text":727,"show_favorite_icon":160},"Become a Partner",{"collection":245,"item":729},{"id":87,"status":5,"sort":10,"url":105,"url_text":104,"show_favorite_icon":160},{"collection":245,"item":731},{"id":95,"status":5,"sort":10,"url":732,"url_text":733,"show_favorite_icon":160},"\u002Ftags\u002Frelease-update","Release Updates",{"navigation_items_group_id":735},{"id":20,"sort":83,"status":5,"name":716,"render_show_all_url":162,"show_all_url_position":161,"show_favorite_icon":160,"show_all_url":101,"represent_as_tile":160,"items":736},[737],{"collection":165,"item":738},{"id":25,"status":5,"sort":10,"show_author_img":160,"blog_post_tag":739,"description":10},"[USE LATEST FEATURED]",{"navigation_items_group_id":741},{"id":41,"sort":71,"status":5,"name":84,"render_show_all_url":162,"show_all_url_position":161,"show_favorite_icon":160,"show_all_url":85,"represent_as_tile":160,"items":742},[743,757,769,780,792,802,813,821],{"collection":174,"item":744},{"id":33,"status":5,"sort":10,"name":745,"item":746},"General",[747],{"collection":748,"item":749},"docs_section",{"id":20,"status":5,"sort":4,"slug":750,"name":745,"icon":751,"featured_icon":754},"general",{"id":752,"title":753},"423101f1-308c-42f1-8dfe-53272019e7cb","Global",{"id":755,"title":756},"0f7ca4cb-dc3b-4f62-b476-132ed5c4931f","General Doc",{"collection":174,"item":758},{"id":37,"status":5,"sort":10,"name":759,"item":760},"CLI",[761],{"collection":748,"item":762},{"id":25,"status":5,"sort":20,"slug":540,"name":759,"icon":763,"featured_icon":766},{"id":764,"title":765},"2f7a4573-8265-4d76-8d55-ce0cc79b8983","Cli",{"id":767,"title":768},"3f76ff7b-0e74-4046-bb03-4ca99c3b66d5","Doc Cat Cli",{"collection":174,"item":770},{"id":53,"status":5,"sort":10,"name":771,"item":772},"API",[773],{"collection":748,"item":774},{"id":107,"status":5,"sort":25,"slug":775,"name":771,"icon":776,"featured_icon":777},"api",{"id":764,"title":765},{"id":778,"title":779},"20866781-e69b-4e01-9456-05437487b75c","API Doc",{"collection":174,"item":781},{"id":782,"status":5,"sort":10,"name":783,"item":784},50,"CDN",[785],{"collection":748,"item":786},{"id":252,"status":5,"sort":33,"slug":787,"name":783,"icon":788,"featured_icon":789},"cdn",{"id":752,"title":753},{"id":790,"title":791},"a8ee9cb7-4e02-41f0-b595-eb518c3085b6","Doc Cat Cdn",{"collection":174,"item":793},{"id":41,"status":5,"sort":10,"name":593,"item":794},[795],{"collection":748,"item":796},{"id":4,"status":5,"sort":37,"slug":592,"name":593,"icon":797,"featured_icon":799},{"id":798,"title":593},"fb8329ae-0c0c-4fbc-bcdc-83f2eeda1039",{"id":800,"title":801},"f7c0dec8-5b42-4943-ab3a-e3665723ad6f","Doc Cat Android",{"collection":174,"item":803},{"id":45,"status":5,"sort":10,"name":600,"item":804},[805],{"collection":748,"item":806},{"id":33,"status":5,"sort":41,"slug":599,"name":600,"icon":807,"featured_icon":810},{"id":808,"title":809},"289e136c-55df-4d51-847b-1782a2308ee4","Ios Blue",{"id":811,"title":812},"2968d3f0-14b5-4427-82e7-f1ea2de17846","Doc Cat Ios",{"collection":174,"item":814},{"id":57,"status":5,"sort":10,"name":671,"item":815},[816],{"collection":748,"item":817},{"id":24,"status":5,"sort":45,"slug":670,"name":671,"icon":818,"featured_icon":819},{"id":764,"title":765},{"id":820,"title":671},"f5e66b5b-a439-47f6-af4a-6ad261de2e87",{"collection":174,"item":822},{"id":61,"status":5,"sort":10,"name":664,"item":823},[824],{"collection":748,"item":825},{"id":280,"status":5,"sort":53,"slug":663,"name":664,"icon":826,"featured_icon":828},{"id":827,"title":664},"3b4b03c7-e131-49e6-a014-c8c2c8a2751e",{"id":829,"title":830},"1dd05c76-e517-4aea-a3d8-49cfddb40056","Strapi Doc",{"id":25,"name":108,"url":109,"url_text":108,"status":5,"groups":832},[],{"id":41,"name":16,"url":834,"url_text":16,"status":5,"groups":835},"\u002Fbook-demo",[],{"status":5,"header_badges":837,"footer_badges":855,"footer_secured_badges":861,"slider_badges":873},[838,845,850],{"id":4,"status":5,"sort":10,"link":839,"title":840,"badgeType":841,"badgeName":842,"metadata":10,"image":843},"https:\u002F\u002Fwww.g2.com\u002Fproducts\u002Flocalazy\u002Freviews","Best meets requirements","Other","G2-winter-24-best-meets-requirments",{"id":844},"8d5dd6fd-6995-4c92-8219-a6ed5ba6acdc",{"id":20,"status":5,"sort":10,"link":839,"title":846,"badgeType":841,"badgeName":847,"metadata":10,"image":848},"Best support","G2-winter-24-best-support",{"id":849},"e59dd465-84d9-45c7-987d-dd4048fd6992",{"id":25,"status":5,"sort":10,"link":839,"title":851,"badgeType":841,"badgeName":852,"metadata":10,"image":853},"Easiest admin","G2-winter-24-easiest-admin",{"id":854},"5f12c4c8-bec0-47cf-87b4-a4502e9f1bd2",[856],{"id":19,"status":5,"sort":10,"link":839,"title":857,"badgeType":841,"badgeName":858,"metadata":10,"image":859},"Users love us","Users love us 25",{"id":860},"232416a5-86ba-4be6-85a5-b88802003d53",[862,868],{"id":15,"status":5,"sort":10,"link":863,"title":864,"badgeType":841,"badgeName":865,"metadata":10,"image":866},"https:\u002F\u002Fsprinto.com\u002Fget-iso-27001\u002F","Sprinto ISO 27001","Sprinto ISO 27001 certificate",{"id":867},"c1c6c06c-6caf-4b78-8a7e-9606c395eb61",{"id":71,"status":5,"sort":10,"link":869,"title":870,"badgeType":841,"badgeName":870,"metadata":10,"image":871},"https:\u002F\u002Fwww.aicpa.org\u002Fsoc4so","SOC certificate",{"id":872},"4f4bc0a6-09a0-41df-8295-8f990d96d941",[874,880,884,888,892,896,901],{"id":37,"status":5,"sort":10,"link":839,"title":875,"badgeType":841,"badgeName":876,"metadata":877,"image":878},"Most likely to recommend","a","",{"id":879},"da055731-3f11-4fc7-96f2-c11c0b84c831",{"id":41,"status":5,"sort":10,"link":839,"title":881,"badgeType":841,"badgeName":10,"metadata":10,"image":882},"EMEA High Performer",{"id":883},"586fb488-81f2-41db-a4e8-ef309e50c0e6",{"id":45,"status":5,"sort":10,"link":839,"title":885,"badgeType":841,"badgeName":10,"metadata":10,"image":886},"Ease of Doing Business With",{"id":887},"bfa26865-430e-463d-9886-cb56a8a8ecf6",{"id":53,"status":5,"sort":10,"link":839,"title":889,"badgeType":841,"badgeName":10,"metadata":10,"image":890},"High Performer",{"id":891},"2cc1ba0f-c23a-4c97-ad4a-a655c97466fe",{"id":57,"status":5,"sort":10,"link":839,"title":893,"badgeType":841,"badgeName":10,"metadata":10,"image":894},"Fastest Implementation",{"id":895},"6f297758-1c95-45f4-a6e9-528319897132",{"id":61,"status":5,"sort":10,"link":897,"title":898,"badgeType":841,"badgeName":10,"metadata":10,"image":899},"https:\u002F\u002Fwww.intercert.com\u002Fservices\u002Fgovernance-risk-compliance\u002Fiso-iec-27001","Intercert ISO 27001",{"id":900},"0521a5f1-e1f2-4da6-a2bc-dc92773f2a7b",{"id":15,"status":5,"sort":10,"link":863,"title":864,"badgeType":841,"badgeName":865,"metadata":10,"image":902},{"id":867},[904,1044,1122,1228,1314,1393,1482,1573,1660,1759],{"id":535,"status":5,"created_on":905,"modified_on":906,"name":607,"slug":606,"description":907,"docs_link":908,"priority":10,"has_sdk":160,"sort":25,"highlighted":160,"in_menu":162,"changefreq":10,"support_type":909,"show_in_spa":162,"is_file_format":160,"meta_title":10,"meta_description":10,"featured_on_web":162,"marketing_subtitle":910,"tile_subtitle":911,"tile_description":912,"is_enterprise":160,"is_popular":160,"primary_use_case":10,"hierarchy_handling":10,"best_for":10,"file_extension":10,"format_code_example":10,"format_to_extension":913,"upload_features":914,"icon":915,"meta_image":916,"primary_tag":918,"tags":934,"integration_methods":952,"recommended_methods":1028,"default_integration_method":1029,"faq_categories":1036,"translations":1042,"platforms_id":535},"2022-03-18T08:09:24.000Z","2025-02-26T13:36:08.000Z","Manage your React app translations with Localazy, a continuous localization tool. React i18n done right. ","\u002Fdocs\u002Fcli\u002Fjson-format","JSON format support","CLI support","react-i18next and more","Use Localazy to translate your React projects with your favorite i18n library.",{"json":647},[],{"id":609},{"id":917},"a97492ec-ee8d-429c-bf66-d0fe59b6e5c9",{"id":37,"status":5,"sort":20,"created_on":919,"label":920,"hidden":160,"icon":921,"translations":923},"2022-03-17T12:23:44.000Z","Web apps",{"id":922},"c9e70e4f-8136-432a-8d82-53c3501a9eb4",[924,926,929,932],{"id":37,"languages_code":925,"label":920},"xxa",{"id":365,"languages_code":927,"label":928},"es","Aplicaciones web",{"id":782,"languages_code":930,"label":931},"cs","Webové aplikace",{"id":565,"languages_code":933,"label":920},"en",[935,942],{"id":37,"status":5,"sort":20,"label":920,"hidden":160,"created_on":919,"icon":936,"translations":937},{"id":922},[938,939,940,941],{"id":37,"languages_code":925,"label":920},{"id":365,"languages_code":927,"label":928},{"id":782,"languages_code":930,"label":931},{"id":565,"languages_code":933,"label":920},{"id":53,"status":5,"sort":57,"label":943,"hidden":160,"created_on":944,"icon":10,"translations":945},"Framework","2022-03-17T12:23:45.000Z",[946,947,949,950],{"id":53,"languages_code":925,"label":943},{"id":500,"languages_code":927,"label":948},"Marco",{"id":451,"languages_code":930,"label":943},{"id":951,"languages_code":933,"label":943},67,[953,965,990,1011],{"id":33,"status":5,"sort":41,"label":771,"alternative_label":10,"slug":775,"description":954,"documentation_link":955,"loc_icon":775,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":956},"Translate strings directly or upload them into your Localazy project.","\u002Fdocs\u002Fapi\u002Fintroduction",[957,958,961,963],{"id":33,"languages_code":925,"label":771,"description":954,"slug":775},{"id":959,"languages_code":927,"label":771,"description":960,"slug":775},40,"Utilice la API para exportar traducciones e importar contenido de\u002Fa Localazy mediante programación.",{"id":782,"languages_code":933,"label":771,"description":962,"slug":775},"Choose between translating strings directly or uploading them into Localazy.",{"id":565,"languages_code":930,"label":771,"description":964,"slug":775},"Pomocí rozhraní API můžete programovaně exportovat překlady a importovat obsah z\u002Fdo Localazy.",{"id":25,"status":5,"sort":37,"label":966,"alternative_label":10,"slug":967,"description":968,"documentation_link":969,"loc_icon":970,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":971},"Create source keys online","add-keys-manually","Add source keys via the web interface online and sync them into your project later.","\u002Fdocs\u002Fgeneral\u002Fimporting-localization-files#add-new-keys-in-ui","vpn-key-outline",[972,973,976,977,980,983,984,985],{"id":25,"languages_code":925,"label":966,"description":968,"slug":967},{"id":291,"languages_code":927,"label":974,"description":975,"slug":10},"Cree claves fuente en línea","Añada claves fuente a través de la interfaz web en línea y sincronícelas posteriormente en su proyecto.",{"id":484,"languages_code":933,"label":966,"description":968,"slug":967},{"id":451,"languages_code":930,"label":978,"description":979,"slug":10},"Vytvářejte zdrojové klíče online","Přidejte zdrojové klíče přes webové rozhraní online a synchronizujte je do svého projektu později.",{"id":402,"languages_code":927,"label":981,"description":982,"slug":967},"Cree claves de origen en línea","Añada claves de origen a través de la interfaz web en línea y sincronícelas posteriormente en su proyecto.",{"id":411,"languages_code":933,"label":966,"description":968,"slug":967},{"id":684,"languages_code":930,"label":978,"description":979,"slug":967},{"id":986,"languages_code":987,"label":988,"description":989,"slug":10},88,"de","Erstellen Sie Quellschlüssel online","Fügen Sie Quellschlüssel über die Weboberfläche online hinzu und synchronisieren Sie sie später mit Ihrem Projekt.",{"id":20,"status":5,"sort":33,"label":991,"alternative_label":992,"slug":993,"description":994,"documentation_link":995,"loc_icon":996,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":997},"Direct file upload","File upload","web-upload","Upload your texts and existing translations in any format directly to start quickly.","\u002Fdocs\u002Fgeneral\u002Fimporting-localization-files","file-upload",[998,999,1002,1003,1007],{"id":20,"languages_code":925,"label":991,"description":994,"slug":993},{"id":218,"languages_code":927,"label":1000,"description":1001,"slug":10},"Carga directa de archivos","Cargue sus textos y traducciones existentes en cualquier formato directamente para empezar rápidamente.",{"id":382,"languages_code":933,"label":991,"description":994,"slug":993},{"id":1004,"languages_code":930,"label":1005,"description":1006,"slug":10},56,"Přímé nahrávání souborů","Nahrajte přímo své texty a stávající překlady v libovolném formátu a začněte pracovat ihned.",{"id":1008,"languages_code":987,"label":1009,"description":1010,"slug":10},85,"Direkter Datei-Upload","Laden Sie Ihre Texte und vorhandenen Übersetzungen in einem beliebigen Format direkt hoch, um schnell loszulegen.",{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1015},"Command Line Interface","The best option for developers that want to make localization an automated part of their workflow.","\u002Fdocs\u002Fcli\u002Fthe-basics",[1016,1017,1020,1021,1024],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},"Interfaz de Línea de Comandos","La mejor opción para los desarrolladores que deseen hacer de la localización una parte automatizada de su flujo de trabajo.",{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},"Rozhraní Příkazového Řádku","Nejlepší volba pro vývojáře, kteří chtějí, aby se lokalizace stala automatizovanou součástí jejich pracovních postupů.",{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},87,"Befehlszeilenschnittstelle","Die beste Option für Entwickler, die die Lokalisierung zu einem automatisierten Teil ihres Arbeitsablaufs machen möchten.",[],{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1030},[1031,1032,1033,1034,1035],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[1037],{"id":1038,"status":5,"created_on":1039,"sort":10,"slug":1040,"label":1041,"on_faq_index":10},76,"2025-02-26T12:44:29.000Z","front-end-common","Front-end Common",[1043],{"languages_code":925,"name":607,"slug":606,"description":907,"support_type":909,"meta_title":10,"meta_description":10,"marketing_subtitle":910,"tile_subtitle":911,"tile_description":912,"id":535,"platforms_id":535,"docs_link":908},{"id":103,"status":5,"created_on":905,"modified_on":1045,"name":614,"slug":613,"description":1046,"docs_link":908,"priority":10,"has_sdk":160,"sort":33,"highlighted":160,"in_menu":162,"changefreq":10,"support_type":909,"show_in_spa":162,"is_file_format":160,"meta_title":10,"meta_description":10,"featured_on_web":162,"marketing_subtitle":910,"tile_subtitle":1047,"tile_description":1048,"is_enterprise":160,"is_popular":160,"primary_use_case":10,"hierarchy_handling":10,"best_for":10,"file_extension":10,"format_code_example":10,"format_to_extension":1049,"upload_features":1050,"icon":1051,"meta_image":1052,"primary_tag":1054,"tags":1061,"integration_methods":1075,"recommended_methods":1106,"default_integration_method":1107,"faq_categories":1114,"translations":1120,"platforms_id":103},"2025-02-26T13:35:41.000Z","Online Vue.js project localization tool for your translation management. Enjoy true continuous localization with Localazy & translate your JSON files automatically.","Vue.js & JSON","Enjoy true continuous localization with Localazy & translate your Vue projects automatically.",{"json":647},[],{"id":616},{"id":1053},"8171b836-d6b8-4354-8db3-34ccd384970d",{"id":37,"status":5,"sort":20,"created_on":919,"label":920,"hidden":160,"icon":1055,"translations":1056},{"id":922},[1057,1058,1059,1060],{"id":37,"languages_code":925,"label":920},{"id":365,"languages_code":927,"label":928},{"id":782,"languages_code":930,"label":931},{"id":565,"languages_code":933,"label":920},[1062,1069],{"id":37,"status":5,"sort":20,"label":920,"hidden":160,"created_on":919,"icon":1063,"translations":1064},{"id":922},[1065,1066,1067,1068],{"id":37,"languages_code":925,"label":920},{"id":365,"languages_code":927,"label":928},{"id":782,"languages_code":930,"label":931},{"id":565,"languages_code":933,"label":920},{"id":53,"status":5,"sort":57,"label":943,"hidden":160,"created_on":944,"icon":10,"translations":1070},[1071,1072,1073,1074],{"id":53,"languages_code":925,"label":943},{"id":500,"languages_code":927,"label":948},{"id":451,"languages_code":930,"label":943},{"id":951,"languages_code":933,"label":943},[1076,1082,1092,1099],{"id":33,"status":5,"sort":41,"label":771,"alternative_label":10,"slug":775,"description":954,"documentation_link":955,"loc_icon":775,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1077},[1078,1079,1080,1081],{"id":33,"languages_code":925,"label":771,"description":954,"slug":775},{"id":959,"languages_code":927,"label":771,"description":960,"slug":775},{"id":782,"languages_code":933,"label":771,"description":962,"slug":775},{"id":565,"languages_code":930,"label":771,"description":964,"slug":775},{"id":25,"status":5,"sort":37,"label":966,"alternative_label":10,"slug":967,"description":968,"documentation_link":969,"loc_icon":970,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1083},[1084,1085,1086,1087,1088,1089,1090,1091],{"id":25,"languages_code":925,"label":966,"description":968,"slug":967},{"id":291,"languages_code":927,"label":974,"description":975,"slug":10},{"id":484,"languages_code":933,"label":966,"description":968,"slug":967},{"id":451,"languages_code":930,"label":978,"description":979,"slug":10},{"id":402,"languages_code":927,"label":981,"description":982,"slug":967},{"id":411,"languages_code":933,"label":966,"description":968,"slug":967},{"id":684,"languages_code":930,"label":978,"description":979,"slug":967},{"id":986,"languages_code":987,"label":988,"description":989,"slug":10},{"id":20,"status":5,"sort":33,"label":991,"alternative_label":992,"slug":993,"description":994,"documentation_link":995,"loc_icon":996,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1093},[1094,1095,1096,1097,1098],{"id":20,"languages_code":925,"label":991,"description":994,"slug":993},{"id":218,"languages_code":927,"label":1000,"description":1001,"slug":10},{"id":382,"languages_code":933,"label":991,"description":994,"slug":993},{"id":1004,"languages_code":930,"label":1005,"description":1006,"slug":10},{"id":1008,"languages_code":987,"label":1009,"description":1010,"slug":10},{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1100},[1101,1102,1103,1104,1105],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[],{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1108},[1109,1110,1111,1112,1113],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[1115,1119],{"id":1004,"status":5,"created_on":1116,"sort":10,"slug":1117,"label":1118,"on_faq_index":10},"2023-03-01T11:21:41.000Z","vue-localization","Vue.js Localization",{"id":1038,"status":5,"created_on":1039,"sort":10,"slug":1040,"label":1041,"on_faq_index":10},[1121],{"languages_code":925,"name":614,"slug":613,"description":1046,"support_type":909,"meta_title":10,"meta_description":10,"marketing_subtitle":910,"tile_subtitle":1047,"tile_description":1048,"id":103,"platforms_id":103,"docs_link":908},{"id":4,"status":5,"created_on":905,"modified_on":1123,"name":627,"slug":626,"description":1124,"docs_link":1125,"priority":10,"has_sdk":160,"sort":41,"highlighted":160,"in_menu":162,"changefreq":10,"support_type":1126,"show_in_spa":162,"is_file_format":160,"meta_title":10,"meta_description":10,"featured_on_web":160,"marketing_subtitle":910,"tile_subtitle":1126,"tile_description":1127,"is_enterprise":160,"is_popular":160,"primary_use_case":1128,"hierarchy_handling":1129,"best_for":1130,"file_extension":1131,"format_code_example":1132,"format_to_extension":1133,"upload_features":1135,"icon":1136,"meta_image":1137,"primary_tag":1139,"tags":1150,"integration_methods":1186,"recommended_methods":1217,"default_integration_method":1218,"faq_categories":1225,"translations":1226,"platforms_id":4},"2025-07-03T14:21:42.000Z","Learn how to use Localazy CLI for localization of Flutter app. Flutter’s ARB format is fully supported including arrays, plurals, and selected context information. No extra configuration is necessary, but you can enable certain features if you want to.","\u002Fdocs\u002Fcli\u002Fflutter-format","ARB format support","Use the Localazy CLI for localization of Flutter projects. Flutter’s ARB format is fully supported.","Supports metadata like placeholders","JSON-like key-value","Flutter apps",".arb","```\n{\n  \"localazy_message\": \"Go international, today. With Localazy.\"\n}\n```",{"arb":1134},"arb",[],{"id":629},{"id":1138},"c2674666-4371-4aa4-9a65-438c7363fafe",{"id":41,"status":5,"sort":53,"created_on":944,"label":1140,"hidden":160,"icon":1141,"translations":1143},"Mobile apps",{"id":1142},"c9c7e1f5-8b59-4016-ae0c-6fba9469cd55",[1144,1145,1147,1149],{"id":41,"languages_code":925,"label":1140},{"id":374,"languages_code":927,"label":1146},"Aplicaciones móviles",{"id":443,"languages_code":930,"label":1148},"Mobilní aplikace",{"id":411,"languages_code":933,"label":1140},[1151,1162,1169,1176],{"id":45,"status":5,"sort":103,"label":1152,"hidden":160,"created_on":944,"icon":1153,"translations":1155},"Desktop & Games",{"id":1154},"65b04533-5b4c-430c-b8a4-ecce84754200",[1156,1157,1159,1161],{"id":45,"languages_code":925,"label":1152},{"id":382,"languages_code":927,"label":1158},"Escritorio y Juegos",{"id":1004,"languages_code":930,"label":1160},"Stolní počítače & Hry",{"id":684,"languages_code":933,"label":1152},{"id":37,"status":5,"sort":20,"label":920,"hidden":160,"created_on":919,"icon":1163,"translations":1164},{"id":922},[1165,1166,1167,1168],{"id":37,"languages_code":925,"label":920},{"id":365,"languages_code":927,"label":928},{"id":782,"languages_code":930,"label":931},{"id":565,"languages_code":933,"label":920},{"id":41,"status":5,"sort":53,"label":1140,"hidden":160,"created_on":944,"icon":1170,"translations":1171},{"id":1142},[1172,1173,1174,1175],{"id":41,"languages_code":925,"label":1140},{"id":374,"languages_code":927,"label":1146},{"id":443,"languages_code":930,"label":1148},{"id":411,"languages_code":933,"label":1140},{"id":223,"status":5,"sort":407,"label":1177,"hidden":162,"created_on":944,"icon":10,"translations":1178},"ARB",[1179,1180,1182,1184],{"id":223,"languages_code":925,"label":1177},{"id":1181,"languages_code":927,"label":1177},120,{"id":1183,"languages_code":930,"label":1177},125,{"id":1185,"languages_code":933,"label":1177},130,[1187,1193,1203,1210],{"id":33,"status":5,"sort":41,"label":771,"alternative_label":10,"slug":775,"description":954,"documentation_link":955,"loc_icon":775,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1188},[1189,1190,1191,1192],{"id":33,"languages_code":925,"label":771,"description":954,"slug":775},{"id":959,"languages_code":927,"label":771,"description":960,"slug":775},{"id":782,"languages_code":933,"label":771,"description":962,"slug":775},{"id":565,"languages_code":930,"label":771,"description":964,"slug":775},{"id":25,"status":5,"sort":37,"label":966,"alternative_label":10,"slug":967,"description":968,"documentation_link":969,"loc_icon":970,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1194},[1195,1196,1197,1198,1199,1200,1201,1202],{"id":25,"languages_code":925,"label":966,"description":968,"slug":967},{"id":291,"languages_code":927,"label":974,"description":975,"slug":10},{"id":484,"languages_code":933,"label":966,"description":968,"slug":967},{"id":451,"languages_code":930,"label":978,"description":979,"slug":10},{"id":402,"languages_code":927,"label":981,"description":982,"slug":967},{"id":411,"languages_code":933,"label":966,"description":968,"slug":967},{"id":684,"languages_code":930,"label":978,"description":979,"slug":967},{"id":986,"languages_code":987,"label":988,"description":989,"slug":10},{"id":20,"status":5,"sort":33,"label":991,"alternative_label":992,"slug":993,"description":994,"documentation_link":995,"loc_icon":996,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1204},[1205,1206,1207,1208,1209],{"id":20,"languages_code":925,"label":991,"description":994,"slug":993},{"id":218,"languages_code":927,"label":1000,"description":1001,"slug":10},{"id":382,"languages_code":933,"label":991,"description":994,"slug":993},{"id":1004,"languages_code":930,"label":1005,"description":1006,"slug":10},{"id":1008,"languages_code":987,"label":1009,"description":1010,"slug":10},{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1211},[1212,1213,1214,1215,1216],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[],{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1219},[1220,1221,1222,1223,1224],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[],[1227],{"languages_code":925,"name":627,"slug":626,"description":1124,"support_type":1126,"meta_title":10,"meta_description":10,"marketing_subtitle":910,"tile_subtitle":1126,"tile_description":1127,"id":4,"platforms_id":4,"docs_link":1125},{"id":570,"status":5,"created_on":1229,"modified_on":1230,"name":641,"slug":640,"description":1231,"docs_link":1232,"priority":10,"has_sdk":160,"sort":19,"highlighted":160,"in_menu":162,"changefreq":10,"support_type":1233,"show_in_spa":162,"is_file_format":160,"meta_title":10,"meta_description":10,"featured_on_web":162,"marketing_subtitle":910,"tile_subtitle":1233,"tile_description":1234,"is_enterprise":160,"is_popular":160,"primary_use_case":10,"hierarchy_handling":10,"best_for":10,"file_extension":10,"format_code_example":10,"format_to_extension":1235,"upload_features":1237,"icon":1238,"meta_image":1239,"primary_tag":1241,"tags":1254,"integration_methods":1269,"recommended_methods":1300,"default_integration_method":1301,"faq_categories":1308,"translations":1312,"platforms_id":570},"2022-03-18T08:09:25.000Z","2024-07-17T07:43:22.000Z","Localize your app with JavaScript localizations and Localazy CLI.\nUpload JavaScript files with strings to Localazy and manage your translations easily. Download translated files back.","\u002Fdocs\u002Fcli\u002Fjavascript-format","JS file support","Translating JS files is easy as a pie with Localazy!",{"js":1236},"js",[],{"id":643},{"id":1240},"2c6699eb-1d80-4772-8268-5597dbcead8c",{"id":61,"status":5,"sort":61,"created_on":944,"label":1242,"hidden":160,"icon":1243,"translations":1245},"Programming Languages",{"id":1244},"8c5836bb-1fbe-49c5-9330-8f931838c457",[1246,1247,1249,1252],{"id":61,"languages_code":925,"label":1242},{"id":468,"languages_code":927,"label":1248},"Lenguajes de Programación",{"id":1250,"languages_code":930,"label":1251},68,"Programovací Jazyky",{"id":1253,"languages_code":933,"label":1242},70,[1255,1262],{"id":37,"status":5,"sort":20,"label":920,"hidden":160,"created_on":919,"icon":1256,"translations":1257},{"id":922},[1258,1259,1260,1261],{"id":37,"languages_code":925,"label":920},{"id":365,"languages_code":927,"label":928},{"id":782,"languages_code":930,"label":931},{"id":565,"languages_code":933,"label":920},{"id":61,"status":5,"sort":61,"label":1242,"hidden":160,"created_on":944,"icon":1263,"translations":1264},{"id":1244},[1265,1266,1267,1268],{"id":61,"languages_code":925,"label":1242},{"id":468,"languages_code":927,"label":1248},{"id":1250,"languages_code":930,"label":1251},{"id":1253,"languages_code":933,"label":1242},[1270,1276,1286,1293],{"id":33,"status":5,"sort":41,"label":771,"alternative_label":10,"slug":775,"description":954,"documentation_link":955,"loc_icon":775,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1271},[1272,1273,1274,1275],{"id":33,"languages_code":925,"label":771,"description":954,"slug":775},{"id":959,"languages_code":927,"label":771,"description":960,"slug":775},{"id":782,"languages_code":933,"label":771,"description":962,"slug":775},{"id":565,"languages_code":930,"label":771,"description":964,"slug":775},{"id":25,"status":5,"sort":37,"label":966,"alternative_label":10,"slug":967,"description":968,"documentation_link":969,"loc_icon":970,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1277},[1278,1279,1280,1281,1282,1283,1284,1285],{"id":25,"languages_code":925,"label":966,"description":968,"slug":967},{"id":291,"languages_code":927,"label":974,"description":975,"slug":10},{"id":484,"languages_code":933,"label":966,"description":968,"slug":967},{"id":451,"languages_code":930,"label":978,"description":979,"slug":10},{"id":402,"languages_code":927,"label":981,"description":982,"slug":967},{"id":411,"languages_code":933,"label":966,"description":968,"slug":967},{"id":684,"languages_code":930,"label":978,"description":979,"slug":967},{"id":986,"languages_code":987,"label":988,"description":989,"slug":10},{"id":20,"status":5,"sort":33,"label":991,"alternative_label":992,"slug":993,"description":994,"documentation_link":995,"loc_icon":996,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1287},[1288,1289,1290,1291,1292],{"id":20,"languages_code":925,"label":991,"description":994,"slug":993},{"id":218,"languages_code":927,"label":1000,"description":1001,"slug":10},{"id":382,"languages_code":933,"label":991,"description":994,"slug":993},{"id":1004,"languages_code":930,"label":1005,"description":1006,"slug":10},{"id":1008,"languages_code":987,"label":1009,"description":1010,"slug":10},{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1294},[1295,1296,1297,1298,1299],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[],{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1302},[1303,1304,1305,1306,1307],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[1309],{"id":464,"status":5,"created_on":1310,"sort":10,"slug":1311,"label":641,"on_faq_index":160},"2022-06-10T09:49:33.000Z","javascript-i18n-localization",[1313],{"languages_code":925,"name":641,"slug":640,"description":1231,"support_type":1233,"meta_title":10,"meta_description":10,"marketing_subtitle":910,"tile_subtitle":1233,"tile_description":1234,"id":570,"platforms_id":570,"docs_link":1232},{"id":19,"status":5,"created_on":1229,"modified_on":1315,"name":648,"slug":647,"description":1316,"docs_link":908,"priority":10,"has_sdk":160,"sort":107,"highlighted":160,"in_menu":162,"changefreq":10,"support_type":1317,"show_in_spa":162,"is_file_format":162,"meta_title":10,"meta_description":10,"featured_on_web":162,"marketing_subtitle":910,"tile_subtitle":1318,"tile_description":1319,"is_enterprise":160,"is_popular":162,"primary_use_case":1320,"hierarchy_handling":1321,"best_for":1322,"file_extension":1323,"format_code_example":1324,"format_to_extension":1325,"upload_features":1326,"icon":1327,"meta_image":1328,"primary_tag":1330,"tags":1341,"integration_methods":1349,"recommended_methods":1380,"default_integration_method":1381,"faq_categories":1388,"translations":1391,"platforms_id":19},"2025-10-14T11:22:29.000Z","Online JSON translator you will love. Comfortable JSON language files editor with powerful CLI under your fingers for continuous localization.","Advanced format support","{Objectively the best}","Translate any JSON files with Localazy, with the best in class plural and array support.","Used for structured data storage and APIs","Uses key-value pairs","Web & mobile apps, APIs",".json","```\n{\n  \"parent\": {\n    \"child\": {\n      \"another_nested_level\": \"All is supported.\"\n    }\n  }\n}\n```",{"json":647},[],{"id":650},{"id":1329},"8d614c20-667e-4bee-a64d-7d7ce9c0ee0a",{"id":33,"status":5,"sort":37,"created_on":944,"label":1331,"hidden":160,"icon":1332,"translations":1334},"File formats",{"id":1333},"ab7c8f89-dcce-45ab-9e21-8605aef289c3",[1335,1336,1338,1340],{"id":33,"languages_code":925,"label":1331},{"id":356,"languages_code":927,"label":1337},"Formatos de archivo",{"id":484,"languages_code":930,"label":1339},"Formáty souborů",{"id":574,"languages_code":933,"label":1331},[1342],{"id":33,"status":5,"sort":37,"label":1331,"hidden":160,"created_on":944,"icon":1343,"translations":1344},{"id":1333},[1345,1346,1347,1348],{"id":33,"languages_code":925,"label":1331},{"id":356,"languages_code":927,"label":1337},{"id":484,"languages_code":930,"label":1339},{"id":574,"languages_code":933,"label":1331},[1350,1356,1366,1373],{"id":33,"status":5,"sort":41,"label":771,"alternative_label":10,"slug":775,"description":954,"documentation_link":955,"loc_icon":775,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1351},[1352,1353,1354,1355],{"id":33,"languages_code":925,"label":771,"description":954,"slug":775},{"id":959,"languages_code":927,"label":771,"description":960,"slug":775},{"id":782,"languages_code":933,"label":771,"description":962,"slug":775},{"id":565,"languages_code":930,"label":771,"description":964,"slug":775},{"id":25,"status":5,"sort":37,"label":966,"alternative_label":10,"slug":967,"description":968,"documentation_link":969,"loc_icon":970,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1357},[1358,1359,1360,1361,1362,1363,1364,1365],{"id":25,"languages_code":925,"label":966,"description":968,"slug":967},{"id":291,"languages_code":927,"label":974,"description":975,"slug":10},{"id":484,"languages_code":933,"label":966,"description":968,"slug":967},{"id":451,"languages_code":930,"label":978,"description":979,"slug":10},{"id":402,"languages_code":927,"label":981,"description":982,"slug":967},{"id":411,"languages_code":933,"label":966,"description":968,"slug":967},{"id":684,"languages_code":930,"label":978,"description":979,"slug":967},{"id":986,"languages_code":987,"label":988,"description":989,"slug":10},{"id":20,"status":5,"sort":33,"label":991,"alternative_label":992,"slug":993,"description":994,"documentation_link":995,"loc_icon":996,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1367},[1368,1369,1370,1371,1372],{"id":20,"languages_code":925,"label":991,"description":994,"slug":993},{"id":218,"languages_code":927,"label":1000,"description":1001,"slug":10},{"id":382,"languages_code":933,"label":991,"description":994,"slug":993},{"id":1004,"languages_code":930,"label":1005,"description":1006,"slug":10},{"id":1008,"languages_code":987,"label":1009,"description":1010,"slug":10},{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1374},[1375,1376,1377,1378,1379],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[],{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1382},[1383,1384,1385,1386,1387],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[1389],{"id":423,"status":5,"created_on":1390,"sort":10,"slug":647,"label":648,"on_faq_index":160},"2022-06-10T10:49:07.000Z",[1392],{"languages_code":925,"name":648,"slug":647,"description":1316,"support_type":1317,"meta_title":10,"meta_description":10,"marketing_subtitle":910,"tile_subtitle":1318,"tile_description":1319,"id":19,"platforms_id":19,"docs_link":908},{"id":24,"status":5,"created_on":1229,"modified_on":1394,"name":655,"slug":654,"description":1395,"docs_link":1396,"priority":10,"has_sdk":160,"sort":24,"highlighted":160,"in_menu":162,"changefreq":10,"support_type":1397,"show_in_spa":160,"is_file_format":162,"meta_title":10,"meta_description":10,"featured_on_web":162,"marketing_subtitle":910,"tile_subtitle":1398,"tile_description":1399,"is_enterprise":160,"is_popular":162,"primary_use_case":1400,"hierarchy_handling":1401,"best_for":1402,"file_extension":1403,"format_code_example":1404,"format_to_extension":1405,"upload_features":1407,"icon":1408,"meta_image":1409,"primary_tag":1411,"tags":1418,"integration_methods":1440,"recommended_methods":1471,"default_integration_method":1472,"faq_categories":1479,"translations":1480,"platforms_id":24},"2025-07-03T14:28:31.000Z","XLIFF is meant for localization. Enjoy Localazy - an online XLIFF editor for your translation management. Integrate your XLIFF files with Localazy to achieve true continuous localization and translate XLIFF files on autopilot.","\u002Fdocs\u002Fcli\u002Fxliff-12-format","File format support","Meant for localization","XLIFF stands for XML Localization Interchange File Format - and Localazy is the best way to manage XLIFF files.","Used in translation pipelines","XML-based exchange format","Translation & software",".xliff","```\n\u003Ctrans-unit id=\"localazy_message\">\n    \u003Csource>Go international, today. With Localazy.\u003C\u002Fsource>\n\u003C\u002Ftrans-unit>\n```",{"xliff":1406},"xlf",[],{"id":657},{"id":1410},"889970a6-cce8-4055-b96a-9c3f292aa67f",{"id":33,"status":5,"sort":37,"created_on":944,"label":1331,"hidden":160,"icon":1412,"translations":1413},{"id":1333},[1414,1415,1416,1417],{"id":33,"languages_code":925,"label":1331},{"id":356,"languages_code":927,"label":1337},{"id":484,"languages_code":930,"label":1339},{"id":574,"languages_code":933,"label":1331},[1419,1426,1433],{"id":33,"status":5,"sort":37,"label":1331,"hidden":160,"created_on":944,"icon":1420,"translations":1421},{"id":1333},[1422,1423,1424,1425],{"id":33,"languages_code":925,"label":1331},{"id":356,"languages_code":927,"label":1337},{"id":484,"languages_code":930,"label":1339},{"id":574,"languages_code":933,"label":1331},{"id":33,"status":5,"sort":37,"label":1331,"hidden":160,"created_on":944,"icon":1427,"translations":1428},{"id":1333},[1429,1430,1431,1432],{"id":33,"languages_code":925,"label":1331},{"id":356,"languages_code":927,"label":1337},{"id":484,"languages_code":930,"label":1339},{"id":574,"languages_code":933,"label":1331},{"id":33,"status":5,"sort":37,"label":1331,"hidden":160,"created_on":944,"icon":1434,"translations":1435},{"id":1333},[1436,1437,1438,1439],{"id":33,"languages_code":925,"label":1331},{"id":356,"languages_code":927,"label":1337},{"id":484,"languages_code":930,"label":1339},{"id":574,"languages_code":933,"label":1331},[1441,1447,1457,1464],{"id":33,"status":5,"sort":41,"label":771,"alternative_label":10,"slug":775,"description":954,"documentation_link":955,"loc_icon":775,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1442},[1443,1444,1445,1446],{"id":33,"languages_code":925,"label":771,"description":954,"slug":775},{"id":959,"languages_code":927,"label":771,"description":960,"slug":775},{"id":782,"languages_code":933,"label":771,"description":962,"slug":775},{"id":565,"languages_code":930,"label":771,"description":964,"slug":775},{"id":25,"status":5,"sort":37,"label":966,"alternative_label":10,"slug":967,"description":968,"documentation_link":969,"loc_icon":970,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1448},[1449,1450,1451,1452,1453,1454,1455,1456],{"id":25,"languages_code":925,"label":966,"description":968,"slug":967},{"id":291,"languages_code":927,"label":974,"description":975,"slug":10},{"id":484,"languages_code":933,"label":966,"description":968,"slug":967},{"id":451,"languages_code":930,"label":978,"description":979,"slug":10},{"id":402,"languages_code":927,"label":981,"description":982,"slug":967},{"id":411,"languages_code":933,"label":966,"description":968,"slug":967},{"id":684,"languages_code":930,"label":978,"description":979,"slug":967},{"id":986,"languages_code":987,"label":988,"description":989,"slug":10},{"id":20,"status":5,"sort":33,"label":991,"alternative_label":992,"slug":993,"description":994,"documentation_link":995,"loc_icon":996,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1458},[1459,1460,1461,1462,1463],{"id":20,"languages_code":925,"label":991,"description":994,"slug":993},{"id":218,"languages_code":927,"label":1000,"description":1001,"slug":10},{"id":382,"languages_code":933,"label":991,"description":994,"slug":993},{"id":1004,"languages_code":930,"label":1005,"description":1006,"slug":10},{"id":1008,"languages_code":987,"label":1009,"description":1010,"slug":10},{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1465},[1466,1467,1468,1469,1470],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[],{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1473},[1474,1475,1476,1477,1478],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[],[1481],{"languages_code":925,"name":655,"slug":654,"description":1395,"support_type":1397,"meta_title":10,"meta_description":10,"marketing_subtitle":910,"tile_subtitle":1398,"tile_description":1399,"id":24,"platforms_id":24,"docs_link":1396},{"id":107,"status":5,"created_on":1483,"modified_on":1483,"name":1484,"slug":1485,"description":1486,"docs_link":1487,"priority":10,"has_sdk":160,"sort":247,"highlighted":160,"in_menu":162,"changefreq":10,"support_type":1397,"show_in_spa":162,"is_file_format":162,"meta_title":10,"meta_description":10,"featured_on_web":160,"marketing_subtitle":910,"tile_subtitle":1488,"tile_description":1489,"is_enterprise":160,"is_popular":162,"primary_use_case":10,"hierarchy_handling":10,"best_for":10,"file_extension":10,"format_code_example":10,"format_to_extension":1490,"upload_features":1491,"icon":1492,"meta_image":1494,"primary_tag":1496,"tags":1503,"integration_methods":1531,"recommended_methods":1562,"default_integration_method":1563,"faq_categories":1570,"translations":1571,"platforms_id":107},"2026-05-21T20:49:09.000Z","PO","po","Edit .po and .pot files with your team, automate translation with AI, and ship faster.","\u002Fdocs\u002Fcli\u002Fpo-format","Seamless integration","Upload, edit and translate PO files with Localazy.",{"po":1485},[],{"id":1493},"25f317fe-0c10-4f0b-92cd-f80d1f3c6a67",{"id":1495},"490022f9-d91a-4e1a-9e79-db374ec997c2",{"id":33,"status":5,"sort":37,"created_on":944,"label":1331,"hidden":160,"icon":1497,"translations":1498},{"id":1333},[1499,1500,1501,1502],{"id":33,"languages_code":925,"label":1331},{"id":356,"languages_code":927,"label":1337},{"id":484,"languages_code":930,"label":1339},{"id":574,"languages_code":933,"label":1331},[1504,1511,1521],{"id":33,"status":5,"sort":37,"label":1331,"hidden":160,"created_on":944,"icon":1505,"translations":1506},{"id":1333},[1507,1508,1509,1510],{"id":33,"languages_code":925,"label":1331},{"id":356,"languages_code":927,"label":1337},{"id":484,"languages_code":930,"label":1339},{"id":574,"languages_code":933,"label":1331},{"id":394,"status":5,"sort":394,"label":1512,"hidden":162,"created_on":944,"icon":10,"translations":1513},"C++",[1514,1515,1517,1519],{"id":394,"languages_code":925,"label":1512},{"id":1516,"languages_code":927,"label":1512},135,{"id":1518,"languages_code":930,"label":1512},140,{"id":1520,"languages_code":933,"label":1512},144,{"id":24,"status":5,"sort":247,"label":1522,"hidden":162,"created_on":944,"icon":10,"translations":1523},"Gettext",[1524,1525,1527,1529],{"id":24,"languages_code":925,"label":1522},{"id":1526,"languages_code":927,"label":1522},101,{"id":1528,"languages_code":930,"label":1522},106,{"id":1530,"languages_code":933,"label":1522},111,[1532,1538,1548,1555],{"id":33,"status":5,"sort":41,"label":771,"alternative_label":10,"slug":775,"description":954,"documentation_link":955,"loc_icon":775,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1533},[1534,1535,1536,1537],{"id":33,"languages_code":925,"label":771,"description":954,"slug":775},{"id":959,"languages_code":927,"label":771,"description":960,"slug":775},{"id":782,"languages_code":933,"label":771,"description":962,"slug":775},{"id":565,"languages_code":930,"label":771,"description":964,"slug":775},{"id":25,"status":5,"sort":37,"label":966,"alternative_label":10,"slug":967,"description":968,"documentation_link":969,"loc_icon":970,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1539},[1540,1541,1542,1543,1544,1545,1546,1547],{"id":25,"languages_code":925,"label":966,"description":968,"slug":967},{"id":291,"languages_code":927,"label":974,"description":975,"slug":10},{"id":484,"languages_code":933,"label":966,"description":968,"slug":967},{"id":451,"languages_code":930,"label":978,"description":979,"slug":10},{"id":402,"languages_code":927,"label":981,"description":982,"slug":967},{"id":411,"languages_code":933,"label":966,"description":968,"slug":967},{"id":684,"languages_code":930,"label":978,"description":979,"slug":967},{"id":986,"languages_code":987,"label":988,"description":989,"slug":10},{"id":20,"status":5,"sort":33,"label":991,"alternative_label":992,"slug":993,"description":994,"documentation_link":995,"loc_icon":996,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1549},[1550,1551,1552,1553,1554],{"id":20,"languages_code":925,"label":991,"description":994,"slug":993},{"id":218,"languages_code":927,"label":1000,"description":1001,"slug":10},{"id":382,"languages_code":933,"label":991,"description":994,"slug":993},{"id":1004,"languages_code":930,"label":1005,"description":1006,"slug":10},{"id":1008,"languages_code":987,"label":1009,"description":1010,"slug":10},{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1556},[1557,1558,1559,1560,1561],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[],{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1564},[1565,1566,1567,1568,1569],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[],[1572],{"languages_code":925,"name":1484,"slug":1485,"description":1486,"support_type":1397,"meta_title":10,"meta_description":10,"marketing_subtitle":910,"tile_subtitle":1488,"tile_description":1489,"id":107,"platforms_id":107,"docs_link":1487},{"id":252,"status":5,"created_on":1229,"modified_on":1574,"name":1575,"slug":1576,"description":1577,"docs_link":1578,"priority":10,"has_sdk":160,"sort":230,"highlighted":160,"in_menu":162,"changefreq":10,"support_type":1397,"show_in_spa":162,"is_file_format":162,"meta_title":10,"meta_description":10,"featured_on_web":160,"marketing_subtitle":910,"tile_subtitle":1579,"tile_description":1580,"is_enterprise":160,"is_popular":160,"primary_use_case":1581,"hierarchy_handling":1582,"best_for":1583,"file_extension":1584,"format_code_example":1585,"format_to_extension":1586,"upload_features":1587,"icon":1588,"meta_image":1589,"primary_tag":1591,"tags":1598,"integration_methods":1616,"recommended_methods":1647,"default_integration_method":1648,"faq_categories":1655,"translations":1658,"platforms_id":252},"2025-10-02T08:32:44.000Z","RESX","resx","Quickly translate and manage your RESX files with Localazy or integrate Localazy with your .NET project and enjoy a fully automated localization process. ","\u002Fdocs\u002Fcli\u002Fresx-format",".NET localization","Translate .NET projects using RESX files and Localazy.","Standard format for Windows\u002F.NET localization","Flat key-value structure",".NET applications",".resx","```\n\u003C?xml version=\"1.0\" encoding=\"utf-8\"?>\n\u003Croot>\n  \u003Cdata name=\"localazy_message\" xml:space=\"preserve\">\n    \u003Cvalue>Go international, today. With Localazy.\u003C\u002Fvalue>\n  \u003C\u002Fdata>\n\u003C\u002Froot>\n```",{"resx":1576},[],{"id":1493},{"id":1590},"1f07bf83-9363-47d6-b21b-a5121c0efaf6",{"id":33,"status":5,"sort":37,"created_on":944,"label":1331,"hidden":160,"icon":1592,"translations":1593},{"id":1333},[1594,1595,1596,1597],{"id":33,"languages_code":925,"label":1331},{"id":356,"languages_code":927,"label":1337},{"id":484,"languages_code":930,"label":1339},{"id":574,"languages_code":933,"label":1331},[1599,1606],{"id":33,"status":5,"sort":37,"label":1331,"hidden":160,"created_on":944,"icon":1600,"translations":1601},{"id":1333},[1602,1603,1604,1605],{"id":33,"languages_code":925,"label":1331},{"id":356,"languages_code":927,"label":1337},{"id":484,"languages_code":930,"label":1339},{"id":574,"languages_code":933,"label":1331},{"id":247,"status":5,"sort":252,"label":1607,"hidden":162,"created_on":944,"icon":10,"translations":1608},"C#",[1609,1610,1612,1614],{"id":247,"languages_code":925,"label":1607},{"id":1611,"languages_code":927,"label":1607},103,{"id":1613,"languages_code":930,"label":1607},108,{"id":1615,"languages_code":933,"label":1607},113,[1617,1623,1633,1640],{"id":33,"status":5,"sort":41,"label":771,"alternative_label":10,"slug":775,"description":954,"documentation_link":955,"loc_icon":775,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1618},[1619,1620,1621,1622],{"id":33,"languages_code":925,"label":771,"description":954,"slug":775},{"id":959,"languages_code":927,"label":771,"description":960,"slug":775},{"id":782,"languages_code":933,"label":771,"description":962,"slug":775},{"id":565,"languages_code":930,"label":771,"description":964,"slug":775},{"id":25,"status":5,"sort":37,"label":966,"alternative_label":10,"slug":967,"description":968,"documentation_link":969,"loc_icon":970,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1624},[1625,1626,1627,1628,1629,1630,1631,1632],{"id":25,"languages_code":925,"label":966,"description":968,"slug":967},{"id":291,"languages_code":927,"label":974,"description":975,"slug":10},{"id":484,"languages_code":933,"label":966,"description":968,"slug":967},{"id":451,"languages_code":930,"label":978,"description":979,"slug":10},{"id":402,"languages_code":927,"label":981,"description":982,"slug":967},{"id":411,"languages_code":933,"label":966,"description":968,"slug":967},{"id":684,"languages_code":930,"label":978,"description":979,"slug":967},{"id":986,"languages_code":987,"label":988,"description":989,"slug":10},{"id":20,"status":5,"sort":33,"label":991,"alternative_label":992,"slug":993,"description":994,"documentation_link":995,"loc_icon":996,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1634},[1635,1636,1637,1638,1639],{"id":20,"languages_code":925,"label":991,"description":994,"slug":993},{"id":218,"languages_code":927,"label":1000,"description":1001,"slug":10},{"id":382,"languages_code":933,"label":991,"description":994,"slug":993},{"id":1004,"languages_code":930,"label":1005,"description":1006,"slug":10},{"id":1008,"languages_code":987,"label":1009,"description":1010,"slug":10},{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1641},[1642,1643,1644,1645,1646],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[],{"id":4,"status":5,"sort":25,"label":1012,"alternative_label":759,"slug":540,"description":1013,"documentation_link":1014,"loc_icon":540,"is_official_plugin":160,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1649},[1650,1651,1652,1653,1654],{"id":4,"languages_code":925,"label":1012,"description":1013,"slug":540},{"id":207,"languages_code":927,"label":1018,"description":1019,"slug":540},{"id":374,"languages_code":933,"label":1012,"description":1013,"slug":540},{"id":443,"languages_code":930,"label":1022,"description":1023,"slug":540},{"id":1025,"languages_code":987,"label":1026,"description":1027,"slug":10},[1656],{"id":356,"status":5,"created_on":1657,"sort":10,"slug":1576,"label":1575,"on_faq_index":10},"2022-08-01T16:44:06.000Z",[1659],{"languages_code":925,"name":1575,"slug":1576,"description":1577,"support_type":1397,"meta_title":10,"meta_description":10,"marketing_subtitle":910,"tile_subtitle":1579,"tile_description":1580,"id":252,"platforms_id":252,"docs_link":1578},{"id":661,"status":5,"created_on":1661,"modified_on":1662,"name":664,"slug":663,"description":1663,"docs_link":1664,"priority":10,"has_sdk":10,"sort":662,"highlighted":160,"in_menu":162,"changefreq":10,"support_type":1665,"show_in_spa":162,"is_file_format":10,"meta_title":10,"meta_description":10,"featured_on_web":162,"marketing_subtitle":1665,"tile_subtitle":1665,"tile_description":1666,"is_enterprise":160,"is_popular":162,"primary_use_case":10,"hierarchy_handling":10,"best_for":10,"file_extension":10,"format_code_example":10,"format_to_extension":10,"upload_features":10,"icon":1667,"meta_image":1668,"primary_tag":1670,"tags":1682,"integration_methods":1726,"recommended_methods":1746,"default_integration_method":1747,"faq_categories":1754,"translations":1757,"platforms_id":661},"2022-06-16T09:08:11.000Z","2026-04-27T18:16:54.000Z","Enjoy seamless integration of Strapi & Localazy and translate your website efficiently. Download the official Localazy plugin in the Strapi Marketplace to get started!","\u002Fdocs\u002Fstrapi\u002Fstrapi-plugin-introduction","Localization Plugin","Localize your Strapi website with our official localization plugin. Easy setup and installation.",{"id":666},{"id":1669},"5995b721-a2ea-487e-ad4b-298a489bbd24",{"id":19,"status":5,"sort":15,"created_on":944,"label":1671,"hidden":160,"icon":1672,"translations":1674},"CMS",{"id":1673},"1172a655-928c-4594-9263-d8afdfd9cd79",[1675,1676,1678,1680],{"id":19,"languages_code":925,"label":1671},{"id":1677,"languages_code":927,"label":1671},73,{"id":1679,"languages_code":930,"label":1671},81,{"id":1681,"languages_code":933,"label":1671},84,[1683,1695,1707,1714],{"id":99,"status":5,"sort":95,"label":1684,"hidden":160,"created_on":944,"icon":10,"translations":1685},"Multilingual SEO",[1686,1687,1690,1693],{"id":99,"languages_code":925,"label":1684},{"id":1688,"languages_code":927,"label":1689},89,"SEO multilingüe",{"id":1691,"languages_code":930,"label":1692},94,"Vícejazyčné SEO",{"id":1694,"languages_code":933,"label":1684},99,{"id":535,"status":5,"sort":535,"label":302,"hidden":160,"created_on":944,"icon":1696,"translations":1698},{"id":1697},"e782ed3f-d954-45ca-8b17-5bf5fe95f30a",[1699,1700,1703,1705],{"id":535,"languages_code":925,"label":302},{"id":1701,"languages_code":927,"label":1702},132,"Comercio electrónico",{"id":1704,"languages_code":930,"label":302},137,{"id":1706,"languages_code":933,"label":302},142,{"id":19,"status":5,"sort":15,"label":1671,"hidden":160,"created_on":944,"icon":1708,"translations":1709},{"id":1673},[1710,1711,1712,1713],{"id":19,"languages_code":925,"label":1671},{"id":1677,"languages_code":927,"label":1671},{"id":1679,"languages_code":930,"label":1671},{"id":1681,"languages_code":933,"label":1671},{"id":20,"status":5,"sort":41,"label":1715,"hidden":160,"created_on":944,"icon":1716,"translations":1718},"Marketing tools",{"id":1717},"098fe6a4-dd2e-42d8-937e-2d8121e6e266",[1719,1720,1723,1725],{"id":20,"languages_code":925,"label":1715},{"id":1721,"languages_code":927,"label":1722},43,"Herramientas de marketing",{"id":431,"languages_code":930,"label":1724},"Marketingové nástroje",{"id":509,"languages_code":933,"label":1715},[1727],{"id":53,"status":5,"sort":10,"label":1728,"alternative_label":1729,"slug":1730,"description":1731,"documentation_link":1732,"loc_icon":1733,"is_official_plugin":162,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1734},"Official Strapi Plugin","Official Plugin","official-strapi-plugin","Our developers maintain an official Strapi localization plugin. Start in a few clicks!","\u002Fdocs\u002Fstrapi","localazy-box",[1735,1736,1739,1740,1743],{"id":53,"languages_code":925,"label":1728,"description":1731,"slug":1730},{"id":1721,"languages_code":927,"label":1737,"description":1738,"slug":10},"Complemento Oficial de Strapi","Nuestros desarrolladores mantienen un complemento oficial de localización de Strapi. ¡Empiece en unos pocos clics!",{"id":500,"languages_code":933,"label":1728,"description":1731,"slug":1730},{"id":574,"languages_code":930,"label":1741,"description":1742,"slug":10},"Oficiál Plugin Strapi","Naši vývojáři udržují a neustále vylepšují oficiální lokalizační plugin Strapi. Začněte několika kliknutími!",{"id":1688,"languages_code":987,"label":1744,"description":1745,"slug":10},"Offizielles Strapi-Plugin","Unsere Entwickler pflegen ein offizielles Strapi-Lokalisierungs-Plugin. Starten Sie mit ein paar Klicks!",[],{"id":53,"status":5,"sort":10,"label":1728,"alternative_label":1729,"slug":1730,"description":1731,"documentation_link":1732,"loc_icon":1733,"is_official_plugin":162,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1748},[1749,1750,1751,1752,1753],{"id":53,"languages_code":925,"label":1728,"description":1731,"slug":1730},{"id":1721,"languages_code":927,"label":1737,"description":1738,"slug":10},{"id":500,"languages_code":933,"label":1728,"description":1731,"slug":1730},{"id":574,"languages_code":930,"label":1741,"description":1742,"slug":10},{"id":1688,"languages_code":987,"label":1744,"description":1745,"slug":10},[1755],{"id":207,"status":5,"created_on":1756,"sort":10,"slug":663,"label":664,"on_faq_index":10},"2022-06-16T09:24:13.000Z",[1758],{"languages_code":925,"name":664,"slug":663,"description":1663,"support_type":1665,"meta_title":10,"meta_description":10,"marketing_subtitle":1665,"tile_subtitle":1665,"tile_description":1666,"id":661,"platforms_id":661,"docs_link":1664},{"id":451,"status":5,"created_on":1229,"modified_on":1760,"name":671,"slug":670,"description":10,"docs_link":1761,"priority":10,"has_sdk":160,"sort":443,"highlighted":160,"in_menu":162,"changefreq":10,"support_type":1762,"show_in_spa":162,"is_file_format":160,"meta_title":10,"meta_description":10,"featured_on_web":162,"marketing_subtitle":1763,"tile_subtitle":1764,"tile_description":1765,"is_enterprise":160,"is_popular":162,"primary_use_case":10,"hierarchy_handling":10,"best_for":10,"file_extension":10,"format_code_example":10,"format_to_extension":10,"upload_features":1766,"icon":1767,"meta_image":1768,"primary_tag":1770,"tags":1783,"integration_methods":1791,"recommended_methods":1804,"default_integration_method":1805,"faq_categories":1811,"translations":1812,"platforms_id":451},"2024-12-20T15:00:22.000Z","\u002Fdocs\u002Fintegrations\u002Fquick-start-figma","plugin","Figma plugin","Localization plugin","Quickly translate your Figma designs and seamlessly reuse the translations during the development of your project.",[],{"id":673},{"id":1769},"7573fed6-8817-4cb4-84d2-be327e6c4891",{"id":25,"status":5,"sort":45,"created_on":944,"label":1771,"hidden":160,"icon":1772,"translations":1774},"Design tools",{"id":1773},"1053d863-b690-4fbd-aadf-dbf2ba282768",[1775,1776,1779,1782],{"id":25,"languages_code":925,"label":1771},{"id":1777,"languages_code":927,"label":1778},42,"Herramientas de diseño",{"id":1780,"languages_code":930,"label":1781},49,"Nástroje pro návrh",{"id":557,"languages_code":933,"label":1771},[1784],{"id":25,"status":5,"sort":45,"label":1771,"hidden":160,"created_on":944,"icon":1785,"translations":1786},{"id":1773},[1787,1788,1789,1790],{"id":25,"languages_code":925,"label":1771},{"id":1777,"languages_code":927,"label":1778},{"id":1780,"languages_code":930,"label":1781},{"id":557,"languages_code":933,"label":1771},[1792],{"id":45,"status":5,"sort":10,"label":1729,"alternative_label":1729,"slug":1793,"description":1794,"documentation_link":85,"loc_icon":1733,"is_official_plugin":162,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1795},"official-plugin","Our developers maintain an official plugin for this integration. Start in a few clicks!",[1796,1797,1800,1801],{"id":45,"languages_code":925,"label":1729,"description":1794,"slug":1793},{"id":280,"languages_code":927,"label":1798,"description":1799,"slug":10},"Plugin Oficial","Nuestros desarrolladores mantienen un plugin oficial para esta integración. ¡Empiece con unos pocos clics!",{"id":271,"languages_code":933,"label":1729,"description":1794,"slug":1793},{"id":464,"languages_code":930,"label":1802,"description":1803,"slug":10},"Oficiální Plugin","Naši vývojáři udržují a neustále vylepšují oficiální plugin pro tuto integraci. Začněte několika kliknutími!",[],{"id":45,"status":5,"sort":10,"label":1729,"alternative_label":1729,"slug":1793,"description":1794,"documentation_link":85,"loc_icon":1733,"is_official_plugin":162,"is_universal_plugin":160,"universal_plugin_vendor":10,"translations":1806},[1807,1808,1809,1810],{"id":45,"languages_code":925,"label":1729,"description":1794,"slug":1793},{"id":280,"languages_code":927,"label":1798,"description":1799,"slug":10},{"id":271,"languages_code":933,"label":1729,"description":1794,"slug":1793},{"id":464,"languages_code":930,"label":1802,"description":1803,"slug":10},[],[1813],{"languages_code":925,"name":671,"slug":670,"description":10,"support_type":1762,"meta_title":10,"meta_description":10,"marketing_subtitle":1763,"tile_subtitle":1764,"tile_description":1765,"id":451,"platforms_id":451,"docs_link":1761},{"id":1815,"owner":1816,"created_by":10,"sort":10,"title":1821,"slug":1822,"modified_on":877,"created_on":1823,"pinned":160,"badge":10,"priority":1824,"main_image":1825,"status":5,"tags":1826,"dictionary":1834,"reading_time":1835,"excerpt":-1,"og_title":877,"og_description":877,"og_image":877},"69d78a237ab1270001eddabf",{"id":1817,"first_name":1818,"last_name":877,"slug":1819,"avatar":1820},"606c233dcf7b6a0001d1da93","Petr Hodný","petr-hodny","https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2023\u002F12\u002Fpetr.png","Translation API: Translate your content on the fly with Localazy AI!","translation-api-translate-your-content-on-the-fly-with-localazy-ai","2026-04-24T09:27:26.000+02:00","0.7","https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2026\u002F04\u002FLocalazy-AI-translation-api.png",[1827,1829,1832],{"id":451,"created_on":1828,"status":5,"label":771,"slug":775,"on_index_page":160},"2026-05-28 16:50:22",{"id":402,"created_on":1828,"status":5,"label":1830,"slug":1831,"on_index_page":162},"AI","ai",{"id":451,"created_on":1828,"status":5,"label":352,"slug":1833,"on_index_page":162},"localization",[],0,{"id":1837,"owner":1838,"created_by":10,"sort":10,"title":1839,"slug":1840,"modified_on":877,"created_on":1841,"pinned":160,"badge":10,"priority":1824,"main_image":1842,"status":5,"tags":1843,"dictionary":1850,"reading_time":1835,"excerpt":-1,"og_title":877,"og_description":877,"og_image":877},"69ea1abb7ab1270001ede902",{"id":1817,"first_name":1818,"last_name":877,"slug":1819,"avatar":1820},"You can just build things. But should you? The build vs. buy debate in localization","build-vs-buy-in-localization-you-can-build-things-but-should-you","2026-05-25T08:43:00.000+02:00","https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2026\u002F05\u002Flocalazy-Vibe-coding.png",[1844,1845,1849],{"id":402,"created_on":1828,"status":5,"label":1830,"slug":1831,"on_index_page":162},{"id":1846,"created_on":1828,"status":5,"label":1847,"slug":1848,"on_index_page":160},646,"Business Development","business-development",{"id":451,"created_on":1828,"status":5,"label":352,"slug":1833,"on_index_page":162},[],{"post":1852,"relatedPosts":2020,"banner":2071},{"id":1853,"owner":1854,"created_by":10,"sort":10,"title":1859,"slug":1860,"modified_on":1861,"created_on":1862,"pinned":162,"badge":10,"priority":1824,"main_image":1863,"status":5,"tags":1864,"dictionary":1872,"reading_time":99,"excerpt":1951,"og_title":877,"og_description":877,"og_image":877,"content":1952,"meta_title":877,"meta_description":877,"canonical":877,"cta":1953,"fullGhostPost":1961},"67c6f7c634f1ce00011a8c97",{"id":1855,"first_name":1856,"last_name":877,"slug":1857,"avatar":1858},"65647a510470910001942870","Eman Abdo","eman-abdo","https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2024\u002F06\u002FEman.png","Can LLMs translate Arabic accurately? We put 8 of them to the test","ai-8-llm-arabic-models-tested-to-translate","2025-08-15T10:44:28.000+02:00","2025-04-30T08:36:00.000+02:00","https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F04\u002FArabicLocal---Challenges-3-1.png",[1865,1868,1869],{"id":402,"created_on":1828,"status":5,"label":1866,"slug":1867,"on_index_page":160},"ChatGPT","chatgpt",{"id":402,"created_on":1828,"status":5,"label":1830,"slug":1831,"on_index_page":162},{"id":451,"created_on":1828,"status":5,"label":1870,"slug":1871,"on_index_page":160},"Locales","locales",[1873,1897,1906,1926,1941],{"id":1874,"status":5,"owner":1875,"created_on":1876,"title":1877,"excerpt":1878,"content":1879,"slug":1880,"meta_title":877,"meta_description":877,"canonical":877,"related_terms":1881},185,"ceebb166-1990-467a-a514-56e71dfcf7a4","2026-05-12T07:09:23.000Z","Large Language Model (LLM)","A neural network that has absorbed and analyzed vast amounts of data and can produce text based on them.","A Large Language Model (LLM) is nothing but a type of Artificial Intelligence (AI) designed to understand and generate human language. These models are trained on vast amounts of text data and utilize machine learning techniques to perform a variety of language-related tasks by synthesizing information and answering to queries.\n\nLLMs are essential in Natural Language Processing (NLP) projects as they help in automating tasks such as text generation, translation, summarization, and more. Each LLM typically represents a sophisticated algorithm capable of producing human-like text based on the patterns it has learned from the training data.\n\n## 🤖 Key points about LLMs: \n\n* Some of the most well-known LLMs include OpenAI's GPT-4, Google's Gemini, and Facebook's RoBERTa.\n* LLMs are used in chatbots, virtual assistants, content creation, and language translation tools.\n* The effectiveness of an LLM depends on the quality and diversity of the text data it has been trained on.\n* The performance of LLMs is measured by their ability to generate coherent, contextually relevant, and grammatically correct text among others.\n* The development and deployment of LLMs raise ethical concerns related to bias, misinformation, and the impact on employment.\n\nUnlike basic language processing tools, LLMs are supposed to understand context, generate nuanced responses, and perform a variety of complex language tasks. Technologies like [Human-Assisted Machine Translation (HAMT)](https:\u002F\u002Flocalazy.com\u002Fdictionary\u002Fhuman-assisted-mt) often integrate LLMs to achieve higher quality and more natural results.\n\n> *Learn more about LLMs and what they are able to do in the translation realm [in our blog](https:\u002F\u002Flocalazy.com\u002Ftags\u002Fai\u002F \"https:\u002F\u002Flocalazy.com\u002Ftags\u002Fai\u002F\").*","llm",[1882,1885,1888,1891,1894],{"id":1883,"slug":1884},169,"ai-powered-translation",{"id":1886,"slug":1887},293,"natural-language-processing-nlp",{"id":1889,"slug":1890},304,"language-modeling",{"id":1892,"slug":1893},439,"tiny-language-model-tlm",{"id":1895,"slug":1896},440,"small-language-model-slm",{"id":1898,"status":5,"owner":1899,"created_on":1900,"title":1901,"excerpt":1902,"content":1903,"slug":1904,"meta_title":10,"meta_description":10,"canonical":10,"related_terms":1905},161,"562275e3-c1f6-48fa-ae16-dc81789ee6e0","2024-05-20T09:52:19.000Z","RTL","Right-to-left scripts, commonly abbreviated as RTL, are writing systems where the writing starts from the right of the page and continues to the left. ","In the software world, RTL is a [locale](https:\u002F\u002Flocalazy.com\u002Fdictionary\u002Flocale) property that needs to be considered during l10n processes.\n\nPopular languages with RTL [text directionality](https:\u002F\u002Flocalazy.com\u002Fdictionary\u002Ftext-directionality) are:\n- Arabic\n- Hebrew\n- Persian (Farsi)\n- Urdu\n- Kurdish\n- Pashto\n- Aramaic\n- Syriac\n\nOriginally, the Chinese, Japanese and Hangul scripts were written right-to-left. Other ancient languages like Old Norse, Old Hungarian, Phoenician and Egyptian hieroglyphs also used RTL directionality.\n\nMost global languages today, like English or Spanish, are LTR – that's why localizing RTL scripts requires extra care and attention. A solid translation management system that includes [the appropiate design tools](https:\u002F\u002Flocalazy.com\u002Fdocs\u002Ffigma\u002Fplugin-introduction#common-i18n-issues-with-designs) to accomodate the differences in directionality and formatting is needed for RTL scripts. This includes post-translation through desktop publishing tools (DTP).","rtl",[],{"id":1907,"status":5,"owner":1875,"created_on":1876,"title":1908,"excerpt":1909,"content":1910,"slug":1911,"meta_title":877,"meta_description":877,"canonical":877,"related_terms":1912},189,"Neural Machine Translation (NMT)","A type of automated translation that uses artificial neural networks to predict the likelihood of a sequence of words, producing translations fast.\n","Neural Machine Translation (NMT) is an advanced approach to automated translation that uses artificial neural networks to predict the likelihood of a sequence of words, typically modeling entire sentences in a single integrated model.\n\nNMT is a key component in the field of [machine translation](https:\u002F\u002Flocalazy.com\u002Fdictionary\u002Fmachine-translation) and represents a significant advancement in [AI-powered translation](https:\u002F\u002Flocalazy.com\u002Fdictionary\u002Fai-powered-translation) technology.\n\nUnlike earlier statistical methods, NMT can capture context and nuances in language, often resulting in more fluent and contextually appropriate translations.\n\n## 🧠 Key points about NMT:\n\n* NMT utilizes deep learning techniques, particularly Recurrent Neural Networks (RNNs) and transformers, to model the translation process. These systems learn to translate directly from source to target language without relying on extensive feature engineering.\n* While initially challenged by rare words, techniques like subword tokenization have improved NMT's ability to handle diverse vocabularies. NMT models can consider the entire input sentence, allowing for better handling of context and long-range dependencies.\n* Training NMT models typically requires significant computational resources and large parallel corpora.\n* Many NMT models incorporate attention mechanisms, allowing the system to focus on relevant parts of the input sentence when generating each word of the translation.\n* Pre-trained NMT models can be adapted to new language pairs or domains, reducing the need for extensive language-specific data.\n* Some models are trained on multiple language pairs simultaneously, enabling zero-shot translation between language pairs not seen during training.\n\nNMT has significantly improved the quality of machine translation across many language pairs and continues to be a primary focus of research and development in the field of AI-powered language technologies.","neural-machine-translation",[1913,1914,1916,1917,1920,1923],{"id":71,"slug":448},{"id":411,"slug":1915},"machine-learning-ml",{"id":1883,"slug":1884},{"id":1918,"slug":1919},186,"statistical-machine-translation",{"id":1921,"slug":1922},187,"example-based-machine-translation",{"id":1924,"slug":1925},188,"hybrid-machine-translation",{"id":1927,"status":5,"owner":1928,"created_on":1929,"title":1930,"excerpt":1931,"content":1932,"slug":1933,"meta_title":877,"meta_description":877,"canonical":877,"related_terms":1934},294,"a5e46ee7-1f50-4f81-ae2d-68a664c76aa2","2026-05-12T07:09:25.000Z","Named Entity Recognition (NER)","A natural language processing technique that automatically finds and labels important names like people, places, or companies in a piece of text.","Named Entity Recognition (NER) is a natural language processing (NLP) technique that identifies and classifies key elements, such as names of people, organizations, locations, dates, and monetary values, within unstructured text. This process transforms raw text into structured data, facilitating tasks like information retrieval, content categorization, and data analysis.\n\nFor example, if a sentence says, *“Apple opened a new office in London,”* NER can spot *“Apple”* as a company and *“London”* as a location. This helps turn messy text into useful, structured data.\n\nNER is helpful in many areas, such as search engines, news filtering, and customer service bots. It lets machines pick out key details from large amounts of text quickly and accurately. To do this well, NER systems often use machine learning, which means they learn from examples to become better at spotting the right words in the right context.\n\nIn localization, NER plays an important role by making sure that names, dates, or other special terms are handled correctly throughout translated content. It helps keep translations accurate, especially when the same word could mean different things in different languages or contexts.\n\nThere are different ways to build NER systems. Some follow fixed rules, while others are trained with data and can adapt over time. Both approaches aim to make it easier for computers to understand language the way humans do.\n\n### 🔍 Core functions of NER: \n\n* **Entity identification**. Detects specific terms in-text, e.g., recognizing *\"Marie Curie\"* as a person or *\"Geneva\"* as a location.\n* **Classification**. Categorizes identified entities into predefined groups like Person, Organization, or Location.\n* **Contextual understanding.** Considers surrounding text to accurately interpret entities, reducing ambiguity.\n\n### 📚 Further reading\n\n* [GeeksforGeeks: Named Entity Recognition](https:\u002F\u002Fwww.geeksforgeeks.org\u002Fnamed-entity-recognition\u002F)\n* [IBM: What Is Named Entity Recognition?](https:\u002F\u002Fwww.ibm.com\u002Fthink\u002Ftopics\u002Fnamed-entity-recognition)\n* [Wikipedia: Named-entity recognition](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FNamed-entity_recognition)","named-entity-recognition-ner",[1935,1938,1939,1940],{"id":1936,"slug":1937},295,"sentence-splitting-segmentation",{"id":411,"slug":1915},{"id":1883,"slug":1884},{"id":1886,"slug":1887},{"id":1886,"status":5,"owner":1928,"created_on":1929,"title":1942,"excerpt":1943,"content":1944,"slug":1887,"meta_title":877,"meta_description":877,"canonical":877,"related_terms":1945},"Natural Language Processing (NLP)","A subfield of artificial intelligence that helps computers understand, interpret, and generate human language.","Natural Language Processing (NLP) is a field of artificial intelligence that enables computers to understand, interpret, and generate human language. It makes this possible by combining linguistics, computer science, and machine learning. It has been revolutionary since it allows us to interact with machines through both text and speech.\n\nNLP is integral to various applications, including chatbots, machine translation, sentiment analysis, and voice assistants. It processes vast amounts of unstructured data from sources such as social media, emails, and documents, extracting meaningful insights and automating tasks that traditionally required human language comprehension.\n\n### 🧠 Core techniques in NLP:\n\n* **Tokenization**: Divides text into individual units, such as words, subwords, or sentences, for easier analysis.\n* **Part-of-Speech (POS) Tagging**: Assigns grammatical categories (e.g., noun, verb) to each word, aiding in syntactic understanding.\n* [**Named Entity Recognition (NER)**](https:\u002F\u002Flocalazy.com\u002Fdictionary\u002Fnamed-entity-recognition-ner): Identifies and classifies entities like names, organizations, locations, and dates within text.\n* **Lemmatization and Stemming**: Reduces words to their base or root forms to standardize variations.\n* **Dependency Parsing**: Analyzes grammatical structure to understand relationships between words in a sentence.\n* **Sentiment Analysis**: Determines the emotional tone or subjective information behind a body of text.\n* **Topic Modeling**: Discovers abstract topics or themes within a collection of documents.\n* **Language Modeling**: Predicts the probability of sequences of words, which is essential for text generation and speech recognition tasks.\n\nAdvancements in deep learning and large language models (LLMs) have significantly enhanced NLP, enabling more accurate and context-aware language understanding. NLP and AI have contributed to the increase of translation quality (and speed), making it easier to achieve better quality localization. \n\n### 📚 **Further reading:**\n\n* [IBM: What is NLP?](https:\u002F\u002Fwww.ibm.com\u002Fthink\u002Ftopics\u002Fnatural-language-processing)\n* [GeeksforGeeks: NLP Overview](https:\u002F\u002Fwww.geeksforgeeks.org\u002Fnatural-language-processing-overview\u002F)\n* [Coursera: NLP Techniques](https:\u002F\u002Fwww.coursera.org\u002Farticles\u002Fnatural-language-processing-techniques)",[1946,1947,1948,1949,1950],{"id":1936,"slug":1937},{"id":411,"slug":1915},{"id":1874,"slug":1880},{"id":1927,"slug":1933},{"id":1889,"slug":1890},"Everyone is using LLMs for almost anything. But are they any good for the nuances of Arabic localization? And what are some worth considering for this scenario? Let's find out.","\u003Cp>Large Language Models are the hottest topic in the localization industry today. While some Natural Language Processing (NLP) tools \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fblog\u002Fthe-great-llm-translation-war-a-comparison-of-the-hottest-ai-models\">are suitable for some languages\u003C\u002Fa>, many are not sufficiently trained for \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fblog\u002Farabic-localization-for-beginners-challenges-opportunities-for-your-global-brand\">Arabic localization\u003C\u002Fa>, which presents unique and complex challenges that stretch their current capabilities.\u003C\u002Fp>\u003Cp>Despite having a massive base of \u003Ca href=\"https:\u002F\u002Fhub.localazy.com\u002Fen\u002Flanguages\u002Far-arabic\">+400 million speakers worldwide\u003C\u002Fa> (spanning the Arab world and other regions, being the official language in more than 26 nations, and having an extensive literary history), Arabic often acts like a language with limited resources when training LLMs for tasks like localization. \u003C\u002Fp>\u003Cp>\u003Cstrong>This isn't because Arabic text is scarce\u003C\u002Fstrong>, but rather due to a mix of linguistic, cultural, and technical issues that make it difficult to generate the data LLMs need to function effectively. That’s why Arabic is considered \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fdictionary\u002Flow-resource-languages\">a low-resource language\u003C\u002Fa>.\u003C\u002Fp>\u003Cblockquote>\u003Cem>Despite being spoken by 400+ million people worldwide, Arabic often acts as a low-resource language for LLMs. This isn't because Arabic text is scarce, but rather because the mix of linguistic, cultural, and technical issues makes it difficult to generate the data needed\u003C\u002Fem>\u003C\u002Fblockquote>\u003Cp>However, because of the current improvements, a light at the end of the seemingly dark tunnel can be seen. \u003Cstrong>Could LLMs improve exponentially to be able to tackle Arabic?\u003C\u002Fstrong> We'll explore this in the article — but first, let's find out why Arabic is so hard to translate for them. \u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002FAdobeStock_264556558.jpeg\" class=\"kg-image\" alt loading=\"lazy\" width=\"2000\" height=\"1333\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002FAdobeStock_264556558.jpeg 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F03\u002FAdobeStock_264556558.jpeg 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1600\u002F2025\u002F03\u002FAdobeStock_264556558.jpeg 1600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw2400\u002F2025\u002F03\u002FAdobeStock_264556558.jpeg 2400w\" sizes=\"(min-width: 720px) 720px\">\u003C\u002Ffigure>\u003Ch2 id=\"why-is-arabic-considered-low-resource-for-llms\">📜 Why is Arabic considered low-resource for LLMs? \u003Ca class=\"markdownit-header-anchor\" href=\"#why-is-arabic-considered-low-resource-for-llms\">🔗\u003C\u002Fa>\u003C\u002Fh2>\u003Cp>There are a few good reasons behind Arabic being considered a low-resource language. They include factors \u003Cstrong>from dialect differences to complicated morphology and \u003C\u002Fstrong>\u003Ca href=\"https:\u002F\u002Fhub.localazy.com\u002Fen\u002Fscripts\u002Farab-arabic\">\u003Cstrong>its unique script\u003C\u002Fstrong>\u003C\u002Fa>. \u003C\u002Fp>\u003Ch3 id=\"dialects-and-variations\">Dialects & variations \u003Ca class=\"markdownit-header-anchor\" href=\"#dialects-and-variations\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Arabic exhibits \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fdictionary\u002Fdiglossia\">diglossia\u003C\u002Fa>, meaning it has two forms: \u003Cstrong>Modern Standard Arabic (MSA)\u003C\u002Fstrong> for formal use and \u003Cstrong>numerous spoken dialects\u003C\u002Fstrong> for daily communication. 🗣️ While MSA text is plentiful, most online content, particularly informal content, uses dialects. \u003C\u002Fp>\u003Cp>\u003Cstrong>LLMs trained mainly on MSA struggle with these dialects,\u003C\u002Fstrong> which are essential for effective localization in fields like marketing, social media, and entertainment. The issue is worsened by \u003Cstrong>the scarcity of transcribed and labeled data for these dialects\u003C\u002Fstrong>, making it hard to train LLMs on them.\u003C\u002Fp>\u003Cp>Arabic dialects aren't a single, uniform language. Some are mutually unintelligible, like \u003Ca href=\"https:\u002F\u002Fhub.localazy.com\u002Fen\u002Flanguages\u002Farz-egyptian-arabic\">Egyptian\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FLevantine_Arabic\">Levantine\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FGulf_Arabic\">Gulf\u003C\u002Fa>, and \u003Ca href=\"https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FMaghrebi_Arabic\">Maghrebi Arabic\u003C\u002Fa>. Spoken Arabic isn't uniform across countries either; it varies regionally. Dialects are further shaped by differences between urban and rural areas and by the impact of foreign languages like French or English. You can think of Arabic and dialects as a sea full of different types of fish. They are all fish, but they are distinct. 🐟\u003C\u002Fp>\u003Cp>This dialectal variation makes it difficult to apply insights or models trained on one dialect to another, since LLMs usually learn from large, general datasets. Because of the complexity of dealing with numerous dialects and the differences between formal and informal language, \u003Cstrong>the abundance of available Arabic data doesn't translate to easy or effective processing for LLMs\u003C\u002Fstrong>. \u003C\u002Fp>\u003Cblockquote>👋 For instance, a simple word like \u003Cem>\"Hello\"\u003C\u002Fem> could be written as \"مرحبًا\" in MSA, but might appear as \"عالسلامة\" in Tunisian Arabic or \"أهلين\" in Levantine Arabic\u003C\u002Fblockquote>\u003Ch3 id=\"complex-morphology\">Complex morphology \u003Ca class=\"markdownit-header-anchor\" href=\"#complex-morphology\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Arabic's complex morphology, where \u003Cstrong>words change form significantly through prefixes, suffixes, infixes, and root patterns\u003C\u002Fstrong>, presents a challenge for LLMs. 🔍 This rich system of word formation, with a single root potentially generating hundreds of different words, adds to the language's complexity for AI. \u003C\u002Fp>\u003Cp>Arabic words frequently stem from a root system, typically three consonants, which form the base meaning. Various patterns of vowels, prefixes, and suffixes are then added to this root to create words with related but distinct meanings. \u003C\u002Fp>\u003Cblockquote>🎨 For example, the root \"r-s-m\" (paint) gives rise to \"rassam\" (painter), and \"rasama\" (he painted). This rich morphology, however, creates ambiguity\u003C\u002Fblockquote>\u003Cp>Because vowels are often left out in written Arabic, a single word can have multiple interpretations, making it difficult for models to understand the correct meaning within a given context.\u003C\u002Fp>\u003Cp>This intricate system makes things even harder for LLMs. They not only have to grasp the root system but also figure out the correct meaning from the context, \u003Cstrong>a process that demands significantly more computing power and advanced model design\u003C\u002Fstrong> than current LLMs often possess.\u003C\u002Fp>\u003Ch3 id=\"script-and-orthography\">Script and orthography \u003Ca class=\"markdownit-header-anchor\" href=\"#script-and-orthography\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>The Arabic script itself presents further hurdles for LLMs. Its right-to-left direction and other unique features set it apart from Latin-based scripts. One key challenge is \u003Cstrong>diacritics\u003C\u002Fstrong>: these marks clarify pronunciation and distinguish words, but are frequently omitted in standard writing, creating ambiguity. \u003C\u002Fp>\u003Cblockquote>👨‍🎨 For instance, the painted form\u003Cstrong> \"رسم\"\u003C\u002Fstrong> could mean \u003Cstrong>\"he painted\" (rasama)\u003C\u002Fstrong>, or \u003Cstrong>\"paint\" (rasm)\u003C\u002Fstrong>, depending on the unwritten diacritics. Also, the way Arabic word endings change based on grammar (like case markers) means a single root can have many different forms\u003C\u002Fblockquote>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002FAdobeStock_492755235.jpeg\" class=\"kg-image\" alt loading=\"lazy\" width=\"2000\" height=\"1333\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002FAdobeStock_492755235.jpeg 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F03\u002FAdobeStock_492755235.jpeg 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1600\u002F2025\u002F03\u002FAdobeStock_492755235.jpeg 1600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw2400\u002F2025\u002F03\u002FAdobeStock_492755235.jpeg 2400w\" sizes=\"(min-width: 720px) 720px\">\u003C\u002Ffigure>\u003Cp>These linguistic features pose specific architectural challenges for LLMs. \u003Cstrong>Standard tokenization methods, often designed for languages like English, aren't well-suited for Arabic\u003C\u002Fstrong>. Because Arabic words are built from roots, splitting them into subwords (a common LLM practice) can obscure important semantic links.\u003C\u002Fp>\u003Cp>In practice, this means that breaking down words from the root\u003Cstrong> \u003C\u002Fstrong>\"ر-س-م\" (r-s-m), such as \"رسم\" (paint), \"رسام\" (painter), and \"مرسم\" (studio), can make it harder for the model to recognize the shared meaning related to \"painting.\"\u003C\u002Fp>\u003Ch3 id=\"data-scarcity-and-imbalance\">Data scarcity and imbalance \u003Ca class=\"markdownit-header-anchor\" href=\"#data-scarcity-and-imbalance\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Despite its large speaker base, Arabic lacks the rich and varied datasets needed for training effective LLMs, especially compared to languages like English. This data scarcity presents several problems:\u003C\u002Fp>\u003Cul>\u003Cli>\u003Cstrong>🏷️ First, there's a shortage of annotated data \u003C\u002Fstrong>for supervised tasks like sentiment analysis, named entity recognition, and translation. The data that \u003Cem>does\u003C\u002Fem> exist often focuses on MSA rather than the more commonly used dialects and informal language. \u003C\u002Fli>\u003Cli>\u003Cstrong>🩺 Second, there are gaps in specific areas\u003C\u002Fstrong>.\u003Cstrong> \u003C\u002Fstrong>Arabic datasets often underrepresent certain regions and specialized fields like medicine or law, hindering a model's ability to perform well in those contexts.\u003C\u002Fli>\u003C\u002Ful>\u003Cp>\u003Cstrong>This data scarcity creates a vicious cycle\u003C\u002Fstrong>. Limited training data results in less effective models, which discourages the use of Arabic language technology. This lack of adoption, in turn, reduces the incentive to invest in better Arabic language support, perpetuating the data shortage.\u003C\u002Fp>\u003Ch3 id=\"design-challenges\">Design challenges \u003Ca class=\"markdownit-header-anchor\" href=\"#design-challenges\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Arabic's right-to-left writing system requires more than just text translation; it demands \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fblog\u002F6-challenges-of-localizing-your-app-to-arabic-and-how-to-solve-them#redesigning-the-ui-and-layout\">\u003Cstrong>a complete redesign of user interfaces\u003C\u002Fstrong>\u003C\u002Fa>. This includes mirroring elements like menus, buttons, text, and images. While LLMs don't manage these layout changes, their output needs to fit well into these reversed interfaces and adapt to specific cultural preferences.\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card kg-card-hascaption\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"1131\" height=\"846\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002Fimage.png 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F03\u002Fimage.png 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage.png 1131w\" sizes=\"(min-width: 720px) 720px\">\u003Cfigcaption>Example:\u003Ca href=\"https:\u002F\u002Fwww.aljazeera.net\u002Ftravel\u002F2025\u002F3\u002F15\u002F%D9%88%D8%AC%D9%87%D8%A9-%D8%B9%D8%B4%D8%A7%D9%82-%D8%A7%D9%84%D8%B3%D9%8A%D8%A7%D8%B1%D8%A7%D8%AA-%D8%B3%D9%8A%D8%A7%D8%AD%D8%A9-%D8%A7%D9%84%D8%B3%D8%B1%D8%B9%D8%A9-%D8%B9%D9%84%D9%89\"> an article on \u003C\u002Fa>Al Jazeera.\u003C\u002Ffigcaption>\u003C\u002Ffigure>\u003Cp>Visual elements like colors and images must be carefully chosen to resonate with the target audience. Achieving this requires tight cooperation between language experts, designers, and developers, significantly increasing the complexity of localization.\u003C\u002Fp>\u003Ch2 id=\"testing-8-llms-on-arabic-translation\">🤖 Testing 8 LLMs on Arabic translation \u003Ca class=\"markdownit-header-anchor\" href=\"#testing-8-llms-on-arabic-translation\">🔗\u003C\u002Fa>\u003C\u002Fh2>\u003Cp>The Arabic language AI technology is promising, but it is developing slowly. The attempts to push further its emerging potential are faced with new ongoing hurdles. \u003Cstrong>These systems have certainly advanced, yet they remain works in progress\u003C\u002Fstrong> rather than final answers to Arabic natural language processing needs — especially when localizing content for Arabic-speaking users.\u003C\u002Fp>\u003Cp>Current Arabic-capable models demonstrate varied capabilities: some excel at formal text generation while stumbling with dialectal variations; others handle basic translation effectively but falter when it comes to cultural nuances. \u003C\u002Fp>\u003Cp>\u003Cstrong>The technology shows promise in structured contexts \u003C\u002Fstrong>like information retrieval and straightforward content creation, but requires human oversight for tasks demanding cultural sensitivity or technical precision.  Here's a summary of some Arabic LLMs' current state, capabilities, strengths, and weaknesses.\u003C\u002Fp>\u003Ch3 id=\"1-gpt-4-multilingual-model\">1. GPT-4 (multilingual model) \u003Ca class=\"markdownit-header-anchor\" href=\"#1-gpt-4-multilingual-model\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>GPT-4 works across multiple languages, including Arabic, with the ability to process and produce text in over 25 languages. The model demonstrates proficiency in generating well-structured sentences in Modern Standard Arabic (MSA) while offering \u003Cstrong>some capability to handle various Arabic dialects\u003C\u002Fstrong>. \u003C\u002Fp>\u003Cp>It serves effectively for conversational tasks, making it \u003Cstrong>suitable for customer support, casual interactions, and information retrieval\u003C\u002Fstrong> in Arabic. Also, it provides reasonable translation services between Arabic and other languages, particularly when working with MSA rather than dialectal variants.\u003C\u002Fp>\u003Cp>The following prompt asks for the translation of \u003Cstrong>يا مساء الزبادي,\u003C\u002Fstrong> an informal way to say \"good evening\" in Egyptian Arabic. The translation is literal and doesn’t convey the playfulness of the greeting:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Flh7-rt.googleusercontent.com\u002Fdocsz\u002FAD_4nXdUr8hJfTLxcNe6Yf374TQyepcj-KNc5RgEZnA7U7KqzZnomdN-vEDZvWsc_wdTk6KU7Q1HNI8qJv-GiCS5ZRZ5Rd1_O6NiDTWUh6v90_Rrtak42M7ZpzdbKKElyjkggdI3HYeJ0YASwhUlp1QL6qM?key=CY7r7ITGL9faXpAb1Za7f00j\" class=\"kg-image\" alt loading=\"lazy\">\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Demonstrates a good general understanding of MSA and some Arabic dialects.\u003C\u002Fli>\u003Cli>Capable of generating creative content in Arabic, such as short stories, articles, and poetry.\u003C\u002Fli>\u003Cli>Maintains contextual coherence in responses, correctly using grammatical elements like tense and subject-object agreement.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Has difficulty with dialectal Arabic. For example, it might misinterpret an Egyptian Arabic phrase or translate it into MSA, losing the intended regional meaning.\u003C\u002Fli>\u003Cli>Struggles with informal, slang-filled Arabic speech.\u003C\u002Fli>\u003Cli>May misinterpret or create ambiguous words due to the common omission of vowels in written Arabic, which can lead to multiple meanings.\u003C\u002Fli>\u003C\u002Ful>\u003Cp>\u003C\u002Fp>\u003Ch3 id=\"2-claude\">2. Claude \u003Ca class=\"markdownit-header-anchor\" href=\"#2-claude\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Claude understands Modern Standard Arabic (MSA) texts accurately and handles complex Arabic grammar correctly. It processes  sentence structures, word forms, and language features properly. \u003Cstrong>The model works consistently well with different types of Arabic text\u003C\u002Fstrong>, including those with challenging grammar elements like dual forms and verb patterns.\u003C\u002Fp>\u003Cp>Here is an example in which Claude explains the translation and the meaning of the sentence:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage-1.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"778\" height=\"241\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002Fimage-1.png 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage-1.png 778w\" sizes=\"(min-width: 720px) 720px\">\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>It performs well in academic and formal settings.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Sometimes it misses cultural nuances.\u003C\u002Fli>\u003Cli>Struggles with dialect-specific expressions.\u003C\u002Fli>\u003Cli>Might produce unnatural-sounding responses in casual conversations.\u003C\u002Fli>\u003Cli>Shows inconsistency in technical terminology, potentially mixing Arabic terms and English loanwords.\u003C\u002Fli>\u003C\u002Ful>\u003Cp>Apart from global, Western models like ChatGPT and Claude, \u003Cstrong>there are several emerging models that focus exclusively on Arabic localization\u003C\u002Fstrong>. Some of these AI tools, designed with cultural and linguistic sensitivity, can leverage 'pronoia' to cultivate positive communication. This means that translations and content are accurate and intentionally crafted to resonate with users on a deeper level. Let's see how they fared in our analysis next.\u003C\u002Fp>\u003Ch3 id=\"3-jais\">3. Jais \u003Ca class=\"markdownit-header-anchor\" href=\"#3-jais\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>\u003Ca href=\"https:\u002F\u002Finceptionai.ai\u002Fjais\u002Findex.html\">Jais\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fmbzuai.ac.ae\u002Fnews\u002Fmeet-jais-the-worlds-most-advanced-arabic-large-language-model-open-sourced-by-g42s-inception\u002F\">released by Inception in 2023,\u003C\u002Fa> was trained on 116 billion Arabic tokens and 279 billion English data tokens. \u003Cstrong>It excels at text generation, creating natural-sounding articles, reports, and stories in Arabic.\u003C\u002Fstrong> It is capable of generating natural Arabic text, sentiment analysis, summarization, \u003Ca href=\"https:\u002F\u002Fwww.techtarget.com\u002Fwhatis\u002Fdefinition\u002Fnamed-entity-recognition-NER\">Named Entity Recognition\u003C\u002Fa>, and Arabic-English translation. Also, it performs well in machine translation between Arabic and other languages.\u003C\u002Fp>\u003Cp>Let's use the same example of مساء الزبادي that we used for ChatGPT. We see that it fails, as it doesn't convey the playfulness of the expression. It says that the translation is literal and needs more context:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage--1--1.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"478\" height=\"389\">\u003C\u002Ffigure>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage--2-.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"563\" height=\"383\">\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>It was trained explicitly on Arabic data, providing a deep understanding of MSA and regional dialects.\u003C\u002Fli>\u003Cli>Highly effective in tasks like text summarization and sentiment analysis for formal Arabic.\u003C\u002Fli>\u003Cli>Designed for business integration: it offers document automation and customer service (chatbots).\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Still struggles with understanding and generating content in various regional Arabic dialects.\u003C\u002Fli>\u003Cli>Tends to produce formal MSA, limiting its usefulness for informal applications.\u003C\u002Fli>\u003Cli>May generate content lacking natural flow or authenticity in creative tasks.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"4-tarjamas-arabic-machine-translation-amt\">4. Tarjama's Arabic Machine Translation (AMT) \u003Ca class=\"markdownit-header-anchor\" href=\"#4-tarjamas-arabic-machine-translation-amt\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>\u003Ca href=\"https:\u002F\u002Ftarjama.com\u002Famt\u002F\">Tarjama's Arabic Machine Translation (AMT)\u003C\u002Fa> system offers \u003Cstrong>business-focused translations with a focus on accuracy and security\u003C\u002Fstrong>. One of its strengths is that it maintains original document formatting and complies with \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fdictionary\u002FISO-27001\">ISO 27001\u003C\u002Fa> standards. AMT can be customized using client-specific data for enhanced precision and uses advanced technology for Arabic translation.\u003C\u002Fp>\u003Cp>The following example is about a proverb coined to indicate the extent of similarity and identity between a mother and her daughter. Regardless of getting a literal translation, it couldn’t differentiate between two Arabic words without diacritics:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Flh7-rt.googleusercontent.com\u002Fdocsz\u002FAD_4nXdv6VrZQMAmJpQ5lTM2TWXtb1agLDmZZ2G95fq2oGMJzndgQzhw1xQbO-lzbr7kcBi0o9uN9gpgskSSAKqTsuJ1RpO413dH-fRc4-N89mdOWLsgnE4tYXOYV7D7IYWi8U-a98ftJu6QU0P0pKm-fA?key=CY7r7ITGL9faXpAb1Za7f00j\" class=\"kg-image\" alt loading=\"lazy\">\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Trained on curated, high-quality data, which contributes to its accuracy.\u003C\u002Fli>\u003Cli>Handles domain-specific terminology (legal, financial, healthcare, etc.).\u003C\u002Fli>\u003Cli>Integrates into existing workflows, supporting various document formats.\u003C\u002Fli>\u003Cli>Incorporates a Human-in-the-Loop approach.\u003C\u002Fli>\u003Cli>Adheres to ISO 27001 standards.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Highly specialized jargon may require human review.\u003C\u002Fli>\u003Cli>Some idiomatic or culturally specific references may need post-editing.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"5-alibaba-clouds-qwen-series-qwen-15-and-qwen-2\">5. Alibaba Cloud's Qwen Series (Qwen 1.5 & Qwen 2) \u003Ca class=\"markdownit-header-anchor\" href=\"#5-alibaba-clouds-qwen-series-qwen-15-and-qwen-2\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Alibaba Cloud's \u003Ca href=\"https:\u002F\u002Fhuggingface.co\u002FQwen\">Qwen series\u003C\u002Fa>, including versions 1.5 and 2, supports over 29 languages including Arabic. \u003Cstrong>It performs well in transcribing and translating, improving speech translation\u003C\u002Fstrong>.\u003C\u002Fp>\u003Cp>Here is a translation example from Qwen where we can see it struggled with context and cultural accuracy:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage--3-.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"1143\" height=\"401\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002Fimage--3-.png 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F03\u002Fimage--3-.png 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage--3-.png 1143w\" sizes=\"(min-width: 720px) 720px\">\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Supports 29 languages, including Arabic.\u003C\u002Fli>\u003Cli>Offers a cost-effective alternative to human translation. \u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>May struggle with cultural nuances and idiomatic expressions.\u003C\u002Fli>\u003Cli>May encounter challenges with specialized or technical jargon.\u003C\u002Fli>\u003Cli>Can face difficulties in comprehending larger or implicit contexts.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"6-arabert\">6. AraBERT \u003Ca class=\"markdownit-header-anchor\" href=\"#6-arabert\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>\u003Ca href=\"https:\u002F\u002Fhuggingface.co\u002Faubmindlab\u002Fbert-base-arabert\">AraBERT \u003C\u002Fa> was \u003Cstrong>specifically designed for the Arabic language based on Google's BERT architecture\u003C\u002Fstrong>. It does a great job in tasks like Named Entity Recognition (NER), part-of-speech tagging, sentiment analysis, and text classification.\u003C\u002Fp>\u003Cp>Now, let's test this model with this \u003Cstrong>source text\u003C\u002Fstrong>: \u003C\u002Fp>\u003Cp>\u003Cem>\"This product will revolutionize the industry.\"\u003C\u002Fem>\u003C\u002Fp>\u003Cp>\u003Cstrong>AraBERT’s Target\u003C\u002Fstrong>: \"سوف تحدث هذه المنتجات ثورة في الصناعة.\"\u003C\u002Fp>\u003Cp>Here, we run into an incorrect word choice or mistranslation. The phrase \"سوف تحدث هذه المنتجات ثورة\" translates to \"these products will revolutionize.\" However, the original source talks about a singular \"product,\" so the plural form (\"المنتجات\") would be a meaning error. \u003C\u002Fp>\u003Cp>The correct translation should be \"سوف تحدث هذه المنتج ثورة في الصناعة.\"\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Efficient for formal MSA.\u003C\u002Fli>\u003Cli>Performs well on document classification, summarization, and question answering.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Poor performance with Arabic dialects.\u003C\u002Fli>\u003Cli>Weak in creative text generation and complex conversations.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"7-camel-camel-labs-arabic-models\">7. CAMeL (CAMeL Lab’s Arabic Models) \u003Ca class=\"markdownit-header-anchor\" href=\"#7-camel-camel-labs-arabic-models\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>\u003Ca href=\"https:\u002F\u002Fhuggingface.co\u002FCAMeL-Lab\u002Fbert-base-arabic-camelbert-da\">CAMeL Lab's Arabic models\u003C\u002Fa> \u003Cstrong>specialize in various dialects: Gulf, Levantine, Egyptian, and Maghrebi\u003C\u002Fstrong>. Their suite of tools, known as CAMeL Tools, offers functionalities like pre-processing, morphological modeling, dialect identification, named entity recognition, and sentiment analysis.\u003C\u002Fp>\u003Cp>What about a translation example? Let's have a look at it. \u003C\u002Fp>\u003Cp>We will use this \u003Cstrong>source text\u003C\u002Fstrong>: \"The match was a real game-changer.\"\u003C\u002Fp>\u003Cp>\u003Cstrong>CAMel’s target translation\u003C\u002Fstrong>: \"كانت المباراة مغيرة حقيقية للعبة.\"\u003C\u002Fp>\u003Cp>We run into a literal translation of an idiomatic expression. The phrase \"game-changer\" is a colloquial expression in English, meaning something that changes the course of an event or situation. CAMeL could have translated this literally as \"مغيرة حقيقية للعبة,\" which would not convey the idiomatic meaning correctly.\u003C\u002Fp>\u003Cp>A better and more accurate translation here would be: \"كانت المباراة نقطة تحول حقيقية.\"\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Fine-tuned for dialectal Arabic.\u003C\u002Fli>\u003Cli>Handles sentiment analysis and code-switching.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Difficulties with contextual coherence in longer texts, especially mixed MSA and dialects.\u003C\u002Fli>\u003Cli>Struggles with highly informal or niche contexts due to being trained on limited data.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"8-tashkeela-qcri\">8. Tashkeela (QCRI) \u003Ca class=\"markdownit-header-anchor\" href=\"#8-tashkeela-qcri\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Qatar Computing Research Institute's (QCRI) \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FAnwarvic\u002FArabic-Tashkeela-Model\">Tashkeela AI model\u003C\u002Fa> \u003Cstrong>adds diacritical marks to Arabic text and performs Named Entity Recognition (NER)\u003C\u002Fstrong> and other Natural Language Processing tasks in MSA. It features a corpus of 75 million fully vocalized words from 97 classical and modern Arabic books, which made possible the development of diacritization systems.\u003C\u002Fp>\u003Cp>Here is an example of using Tashkeela to add diacritization to Arabic texts to make them clearer:\u003C\u002Fp>\u003Cul>\u003Cli>\u003Cstrong>Arabic text without diacritics\u003C\u002Fstrong>:\u003Cbr>\"ذهب الولد إلى المدرسة\"\u003C\u002Fli>\u003Cli>\u003Cstrong>Arabic text with diacritics\u003C\u002Fstrong>:\u003Cbr>\"ذَهَبَ اَلْوَلَدُ إِلَى اَلْمَدْرَسَةِ\"\u003C\u002Fli>\u003Cli>\u003Cstrong>English translation\u003C\u002Fstrong>: \u003Cbr>\"The boy went to school.\"\u003C\u002Fli>\u003C\u002Ful>\u003Cblockquote>\u003Cem>\u003Cstrong>Note:\u003C\u002Fstrong>\u003C\u002Fem> This model is used to add diacritics to clarify the meaning and provide an accurate translation. It \u003Cem>doesn't \u003C\u002Fem>provide translation. Its accuracy is continuously increasing, becoming more and more reliable.\u003C\u002Fblockquote>\u003Cfigure class=\"kg-card kg-image-card kg-card-hascaption\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F04\u002FExample-of-Tashkeela.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"1164\" height=\"281\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F04\u002FExample-of-Tashkeela.png 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F04\u002FExample-of-Tashkeela.png 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F04\u002FExample-of-Tashkeela.png 1164w\" sizes=\"(min-width: 720px) 720px\">\u003Cfigcaption>Tashkeela-Model (\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FAnwarvic\u002FArabic-Tashkeela-Model\">GitHub\u003C\u002Fa>)\u003C\u002Ffigcaption>\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>It reduces ambiguity through diacritics.\u003C\u002Fli>\u003Cli>With a focus on MSA, it performs well with standardized Arabic.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Struggles with non-standard Arabic.\u003C\u002Fli>\u003Cli>Since it's limited to diacritics, it offers limited utility for deeper text understanding or generation.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"my-picks-as-a-translator\">☝️ My picks as a translator \u003Ca class=\"markdownit-header-anchor\" href=\"#my-picks-as-a-translator\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Although all models are still a work in progress, \u003Cstrong>Claude\u003C\u002Fstrong> is my personal pick for Arabic localization tasks. The LLM by Anthropic is advancing constantly — it understands cultural references and considers the nuances of the language. It can also:\u003C\u002Fp>\u003Cul>\u003Cli>🚩 Detect literal translation and help revise long texts, suggesting better options when asked. \u003C\u002Fli>\u003Cli>✏️ Help with different types of content, providing accurate equivalents, and considering context.\u003C\u002Fli>\u003Cli>💬 Remind the user of the different word meanings that could be used depending on the context.\u003C\u002Fli>\u003C\u002Ful>\u003Cp>For simple translation, \u003Cstrong>ChatGPT \u003C\u002Fstrong>is also a good option, considering how it understands different dialects. It can help identify the nature of the source and suggest a suitable translation. However, it might skip some paragraphs or lines.\u003C\u002Fp>\u003Cp>Finally, \u003Cstrong>AMT\u003C\u002Fstrong> and \u003Cstrong>Qwen\u003C\u002Fstrong> are suitable options if you're dealing with business and customer service-related content, but keep in mind that they don't always get dialects correctly. They are progressing, but until now, I haven't found a better option than Claude.\u003C\u002Fp>\u003Cblockquote>🎙️ How realistic are agentic workflows in localization, and are we ready to implement them? \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fblog\u002Fmost-localization-teams-arent-ready-for-ai-workflows-bridging-the-gap-s02-ep08\">Listen to our Bridging the Gap podcast episode with Julia Díez\u003C\u002Fa> for a deep dive into it.\u003C\u002Fblockquote>\u003Ch2 id=\"can-ai-deal-with-my-arabic-content\">🤷 Can AI deal with my Arabic content? \u003Ca class=\"markdownit-header-anchor\" href=\"#can-ai-deal-with-my-arabic-content\">🔗\u003C\u002Fa>\u003C\u002Fh2>\u003Cp>Now, you might ask, \u003Cstrong>\"How can I know if the LLM I use suits my content translation needs?\u003C\u002Fstrong>\". Well, that's a fair question. Here are some factors you should take into consideration to assess this:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002FAdobeStock_1102016803.jpeg\" class=\"kg-image\" alt loading=\"lazy\" width=\"2000\" height=\"1334\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002FAdobeStock_1102016803.jpeg 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F03\u002FAdobeStock_1102016803.jpeg 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1600\u002F2025\u002F03\u002FAdobeStock_1102016803.jpeg 1600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw2400\u002F2025\u002F03\u002FAdobeStock_1102016803.jpeg 2400w\" sizes=\"(min-width: 720px) 720px\">\u003C\u002Ffigure>\u003Ch3 id=\"1-industry-specificity\">1. Industry specificity \u003Ca class=\"markdownit-header-anchor\" href=\"#1-industry-specificity\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Check if the model is trained on domain-specific data (legal, medical, etc.) and if your content is technical\u002Fspecialized. You need to determine how standardized the terminology is.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Look for fine-tuned models (AraBERT, CAMeL) or industry benchmarks.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"2-target-audienceregion\">2. Target audience\u002Fregion \u003Ca class=\"markdownit-header-anchor\" href=\"#2-target-audienceregion\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Ensure the model handles dialectal variations (Egyptian, Gulf, etc.). Check which Arabic variants are used and if there are multiple dialects needed or a required formality level.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Check dialect support (CAMeL) vs. MSA defaults.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"3-user-experience\">3. User experience \u003Ca class=\"markdownit-header-anchor\" href=\"#3-user-experience\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Verify contextual relevance and coherent dialogue in conversations.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Test conversational flows (GPT-4) and dialect-specific training.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"4-informalsocial-media-language\">4. Informal\u002Fsocial media language \u003Ca class=\"markdownit-header-anchor\" href=\"#4-informalsocial-media-language\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>If your content includes more informal language, check how the model handles slang, abbreviations, and code-switching.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Test with social media samples; note limitations of MSA-focused models.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"5-nuanced-sentiment-analysis\">5. Nuanced sentiment analysis \u003Ca class=\"markdownit-header-anchor\" href=\"#5-nuanced-sentiment-analysis\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Test how the LLM handles complex emotions, like sarcasm and ambiguity.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Run tests with ambiguous\u002Femotional phrases; consider AraBERT or CAMeL.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"6-undiacritized-arabic\">6. Undiacritized Arabic \u003Ca class=\"markdownit-header-anchor\" href=\"#6-undiacritized-arabic\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Test the model's ability to handle text without vowel markings.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Use Tashkeela for diacritization, and assess general LLM context inference.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"7-multilingualcode-switched-content\">7. Multilingual\u002Fcode-switched content \u003Ca class=\"markdownit-header-anchor\" href=\"#7-multilingualcode-switched-content\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Check for fluid language switching and bilingual content processing.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Test with Arabic-English or other mixed language content.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"8-content-formatstyle-variability\">8. Content format\u002Fstyle variability \u003Ca class=\"markdownit-header-anchor\" href=\"#8-content-formatstyle-variability\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Ensure the model handles different registers and contexts.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Test with diverse text types (news, conversations) for contextual appropriateness.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"9-real-time-processing\">9. Real-time processing \u003Ca class=\"markdownit-header-anchor\" href=\"#9-real-time-processing\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>For time-sensitive applications, evaluate processing times of the different options.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong> Test response time and processing speed, especially in real-time scenarios.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"10-accuracy-vs-functionality\">10. Accuracy vs. functionality \u003Ca class=\"markdownit-header-anchor\" href=\"#10-accuracy-vs-functionality\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Determine the required accuracy level for your specific use case.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Check accuracy rates across various tasks (translation, summarization).\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"11-performance-evaluation\">11. Performance evaluation \u003Ca class=\"markdownit-header-anchor\" href=\"#11-performance-evaluation\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Monitor overall tendencies like accuracy, dialect handling, and technical term consistency.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Use side-by-side human review to score translations on clarity and accuracy.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"12-error-analysis\">12. Error analysis \u003Ca class=\"markdownit-header-anchor\" href=\"#12-error-analysis\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Finally, document dialect mismatches, technical term confusion, formality issues, and cultural misunderstandings. Evaluate error significance (safety, brand, user experience, legal). Decide how much human oversight is needed.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Track the top error types with color-coded severity ratings (red\u002Fyellow\u002Fgreen).\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"13-final-check-sample-data-testing\">13. Final check: sample data testing \u003Ca class=\"markdownit-header-anchor\" href=\"#13-final-check-sample-data-testing\">🔗\u003C\u002Fa>\u003C\u002Fh3>\u003Cp>Crucially, always test with real sample data to identify specific strengths and weaknesses. Create representative samples with dialects, technical terms, formality variations, numbers, and dates.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Test with a mix of everyday and technical Arabic text samples.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cp>After considering all of these aspects, you can go ahead and use the AI model of your choice for a while to see how it performs. But if you're still unsure about using AI in your industry, this would be \u003Cstrong>a quick and brief categorization\u003C\u002Fstrong> that tells you if it would make sense to use AI in your industry or not: \u003C\u002Fp>\u003Cul>\u003Cli>\u003Cstrong>🟢 Safe for AI\u003C\u002Fstrong>:\u003Cstrong> \u003C\u002Fstrong>General MSA information, basic queries, non-critical content, and standard communication.\u003C\u002Fli>\u003Cli>\u003Cstrong>🟠 Requires careful monitoring\u003C\u002Fstrong>:\u003Cstrong> \u003C\u002Fstrong>Mixed dialects, semi-technical documentation, marketing, education.\u003C\u002Fli>\u003Cli>\u003Cstrong>🔴 Not ready for AI\u003C\u002Fstrong>:\u003Cstrong> \u003C\u002Fstrong>Legal documents, medical instructions, safety-critical information, creative content, and complex manuals.\u003C\u002Fli>\u003C\u002Ful>\u003Ch2 id=\"conclusion-is-it-worth-it-to-use-ai-to-localize-to-arabic\">👀 Conclusion: Is it worth it to use AI to localize to Arabic? \u003Ca class=\"markdownit-header-anchor\" href=\"#conclusion-is-it-worth-it-to-use-ai-to-localize-to-arabic\">🔗\u003C\u002Fa>\u003C\u002Fh2>\u003Cp>\u003Cstrong>AI is a valuable assistant for Arabic content, but has limitations\u003C\u002Fstrong>. It still makes mistakes regardless of what model you choose, especially when the material is nuanced and contains specialized vocabulary. \u003C\u002Fp>\u003Cp>Arabic's complexity requires contextual understanding that AI alone cannot fully provide. While AI can translate and generate content to some point, \u003Cstrong>human oversight is what controls the cultural relevance, proper right-to-left layouts, and brand alignment\u003C\u002Fstrong>. Dialectal Arabic particularly challenges AI systems, often necessitating human adaptation. \u003C\u002Fp>\u003Cp>Critical documents like legal or medical texts absolutely require human review. If you're dealing with that type of sensitive material (or with technical or very nuanced content in Arabic), proofreading is still highly important. We're here to help — get in touch with expert Arabic translators from our \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fdocs\u002Fgeneral\u002Fcontinuous-localization-team\">Continuous Localization Team\u003C\u002Fa>, or begin experimenting with Localazy on your own for your Arabic-speaking projects.\u003C\u002Fp>",{"id":570,"status":5,"created_on":1954,"modified_on":1955,"icon":1956,"header":1957,"description":1958,"button_label":1959,"link":1960},"2022-11-09T12:44:15.000Z","2023-01-19T11:53:19.000Z","pub","Order Professional Translations 💎","Localazy can take complete care of your translation process. Forget the hassle of managing translation projects forever. All you need to do is choose the language and service. Try it now!","Order now","my\u002Fvirtual-translator",{"slug":1860,"id":1853,"uuid":1962,"title":1859,"html":1963,"comment_id":1853,"feature_image":1863,"featured":162,"visibility":1964,"email_recipient_filter":1965,"created_at":1966,"updated_at":1861,"published_at":1862,"custom_excerpt":1951,"codeinjection_head":10,"codeinjection_foot":10,"custom_template":10,"canonical_url":10,"authors":1967,"tags":1972,"primary_author":2017,"primary_tag":2018,"url":2019,"excerpt":1951,"reading_time":83,"access":162,"send_email_when_published":160,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"email_subject":10,"frontmatter":10,"dictionary":1872,"cta":1953,"plainTags":1864},"ec6defc8-cb2f-4c4f-8108-45d074bfad2f","\u003Cp>Large Language Models are the hottest topic in the localization industry today. While some Natural Language Processing (NLP) tools \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fblog\u002Fthe-great-llm-translation-war-a-comparison-of-the-hottest-ai-models\">are suitable for some languages\u003C\u002Fa>, many are not sufficiently trained for \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fblog\u002Farabic-localization-for-beginners-challenges-opportunities-for-your-global-brand\">Arabic localization\u003C\u002Fa>, which presents unique and complex challenges that stretch their current capabilities.\u003C\u002Fp>\u003Cp>Despite having a massive base of \u003Ca href=\"https:\u002F\u002Fhub.localazy.com\u002Fen\u002Flanguages\u002Far-arabic\">+400 million speakers worldwide\u003C\u002Fa> (spanning the Arab world and other regions, being the official language in more than 26 nations, and having an extensive literary history), Arabic often acts like a language with limited resources when training LLMs for tasks like localization. \u003C\u002Fp>\u003Cp>\u003Cstrong>This isn't because Arabic text is scarce\u003C\u002Fstrong>, but rather due to a mix of linguistic, cultural, and technical issues that make it difficult to generate the data LLMs need to function effectively. That’s why Arabic is considered \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fdictionary\u002Flow-resource-languages\">a low-resource language\u003C\u002Fa>.\u003C\u002Fp>\u003Cblockquote>\u003Cem>Despite being spoken by 400+ million people worldwide, Arabic often acts as a low-resource language for LLMs. This isn't because Arabic text is scarce, but rather because the mix of linguistic, cultural, and technical issues makes it difficult to generate the data needed\u003C\u002Fem>\u003C\u002Fblockquote>\u003Cp>However, because of the current improvements, a light at the end of the seemingly dark tunnel can be seen. \u003Cstrong>Could LLMs improve exponentially to be able to tackle Arabic?\u003C\u002Fstrong> We'll explore this in the article — but first, let's find out why Arabic is so hard to translate for them. \u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002FAdobeStock_264556558.jpeg\" class=\"kg-image\" alt loading=\"lazy\" width=\"2000\" height=\"1333\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002FAdobeStock_264556558.jpeg 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F03\u002FAdobeStock_264556558.jpeg 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1600\u002F2025\u002F03\u002FAdobeStock_264556558.jpeg 1600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw2400\u002F2025\u002F03\u002FAdobeStock_264556558.jpeg 2400w\" sizes=\"(min-width: 720px) 720px\">\u003C\u002Ffigure>\u003Ch2 id=\"%F0%9F%93%9C-why-is-arabic-considered-low-resource-for-llms\">📜 Why is Arabic considered low-resource for LLMs?\u003C\u002Fh2>\u003Cp>There are a few good reasons behind Arabic being considered a low-resource language. They include factors \u003Cstrong>from dialect differences to complicated morphology and \u003C\u002Fstrong>\u003Ca href=\"https:\u002F\u002Fhub.localazy.com\u002Fen\u002Fscripts\u002Farab-arabic\">\u003Cstrong>its unique script\u003C\u002Fstrong>\u003C\u002Fa>. \u003C\u002Fp>\u003Ch3 id=\"dialects-variations\">Dialects & variations\u003C\u002Fh3>\u003Cp>Arabic exhibits \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fdictionary\u002Fdiglossia\">diglossia\u003C\u002Fa>, meaning it has two forms: \u003Cstrong>Modern Standard Arabic (MSA)\u003C\u002Fstrong> for formal use and \u003Cstrong>numerous spoken dialects\u003C\u002Fstrong> for daily communication. 🗣️ While MSA text is plentiful, most online content, particularly informal content, uses dialects. \u003C\u002Fp>\u003Cp>\u003Cstrong>LLMs trained mainly on MSA struggle with these dialects,\u003C\u002Fstrong> which are essential for effective localization in fields like marketing, social media, and entertainment. The issue is worsened by \u003Cstrong>the scarcity of transcribed and labeled data for these dialects\u003C\u002Fstrong>, making it hard to train LLMs on them.\u003C\u002Fp>\u003Cp>Arabic dialects aren't a single, uniform language. Some are mutually unintelligible, like \u003Ca href=\"https:\u002F\u002Fhub.localazy.com\u002Fen\u002Flanguages\u002Farz-egyptian-arabic\">Egyptian\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FLevantine_Arabic\">Levantine\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FGulf_Arabic\">Gulf\u003C\u002Fa>, and \u003Ca href=\"https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FMaghrebi_Arabic\">Maghrebi Arabic\u003C\u002Fa>. Spoken Arabic isn't uniform across countries either; it varies regionally. Dialects are further shaped by differences between urban and rural areas and by the impact of foreign languages like French or English. You can think of Arabic and dialects as a sea full of different types of fish. They are all fish, but they are distinct. 🐟\u003C\u002Fp>\u003Cp>This dialectal variation makes it difficult to apply insights or models trained on one dialect to another, since LLMs usually learn from large, general datasets. Because of the complexity of dealing with numerous dialects and the differences between formal and informal language, \u003Cstrong>the abundance of available Arabic data doesn't translate to easy or effective processing for LLMs\u003C\u002Fstrong>. \u003C\u002Fp>\u003Cblockquote>👋 For instance, a simple word like \u003Cem>\"Hello\"\u003C\u002Fem> could be written as \"مرحبًا\" in MSA, but might appear as \"عالسلامة\" in Tunisian Arabic or \"أهلين\" in Levantine Arabic\u003C\u002Fblockquote>\u003Ch3 id=\"complex-morphology\">Complex morphology\u003C\u002Fh3>\u003Cp>Arabic's complex morphology, where \u003Cstrong>words change form significantly through prefixes, suffixes, infixes, and root patterns\u003C\u002Fstrong>, presents a challenge for LLMs. 🔍 This rich system of word formation, with a single root potentially generating hundreds of different words, adds to the language's complexity for AI. \u003C\u002Fp>\u003Cp>Arabic words frequently stem from a root system, typically three consonants, which form the base meaning. Various patterns of vowels, prefixes, and suffixes are then added to this root to create words with related but distinct meanings. \u003C\u002Fp>\u003Cblockquote>🎨 For example, the root \"r-s-m\" (paint) gives rise to \"rassam\" (painter), and \"rasama\" (he painted). This rich morphology, however, creates ambiguity\u003C\u002Fblockquote>\u003Cp>Because vowels are often left out in written Arabic, a single word can have multiple interpretations, making it difficult for models to understand the correct meaning within a given context.\u003C\u002Fp>\u003Cp>This intricate system makes things even harder for LLMs. They not only have to grasp the root system but also figure out the correct meaning from the context, \u003Cstrong>a process that demands significantly more computing power and advanced model design\u003C\u002Fstrong> than current LLMs often possess.\u003C\u002Fp>\u003Ch3 id=\"script-and-orthography\">Script and orthography\u003C\u002Fh3>\u003Cp>The Arabic script itself presents further hurdles for LLMs. Its right-to-left direction and other unique features set it apart from Latin-based scripts. One key challenge is \u003Cstrong>diacritics\u003C\u002Fstrong>: these marks clarify pronunciation and distinguish words, but are frequently omitted in standard writing, creating ambiguity. \u003C\u002Fp>\u003Cblockquote>👨‍🎨 For instance, the painted form\u003Cstrong> \"رسم\"\u003C\u002Fstrong> could mean \u003Cstrong>\"he painted\" (rasama)\u003C\u002Fstrong>, or \u003Cstrong>\"paint\" (rasm)\u003C\u002Fstrong>, depending on the unwritten diacritics. Also, the way Arabic word endings change based on grammar (like case markers) means a single root can have many different forms\u003C\u002Fblockquote>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002FAdobeStock_492755235.jpeg\" class=\"kg-image\" alt loading=\"lazy\" width=\"2000\" height=\"1333\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002FAdobeStock_492755235.jpeg 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F03\u002FAdobeStock_492755235.jpeg 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1600\u002F2025\u002F03\u002FAdobeStock_492755235.jpeg 1600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw2400\u002F2025\u002F03\u002FAdobeStock_492755235.jpeg 2400w\" sizes=\"(min-width: 720px) 720px\">\u003C\u002Ffigure>\u003Cp>These linguistic features pose specific architectural challenges for LLMs. \u003Cstrong>Standard tokenization methods, often designed for languages like English, aren't well-suited for Arabic\u003C\u002Fstrong>. Because Arabic words are built from roots, splitting them into subwords (a common LLM practice) can obscure important semantic links.\u003C\u002Fp>\u003Cp>In practice, this means that breaking down words from the root\u003Cstrong> \u003C\u002Fstrong>\"ر-س-م\" (r-s-m), such as \"رسم\" (paint), \"رسام\" (painter), and \"مرسم\" (studio), can make it harder for the model to recognize the shared meaning related to \"painting.\"\u003C\u002Fp>\u003Ch3 id=\"data-scarcity-and-imbalance\">Data scarcity and imbalance\u003C\u002Fh3>\u003Cp>Despite its large speaker base, Arabic lacks the rich and varied datasets needed for training effective LLMs, especially compared to languages like English. This data scarcity presents several problems:\u003C\u002Fp>\u003Cul>\u003Cli>\u003Cstrong>🏷️ First, there's a shortage of annotated data \u003C\u002Fstrong>for supervised tasks like sentiment analysis, named entity recognition, and translation. The data that \u003Cem>does\u003C\u002Fem> exist often focuses on MSA rather than the more commonly used dialects and informal language. \u003C\u002Fli>\u003Cli>\u003Cstrong>🩺 Second, there are gaps in specific areas\u003C\u002Fstrong>.\u003Cstrong> \u003C\u002Fstrong>Arabic datasets often underrepresent certain regions and specialized fields like medicine or law, hindering a model's ability to perform well in those contexts.\u003C\u002Fli>\u003C\u002Ful>\u003Cp>\u003Cstrong>This data scarcity creates a vicious cycle\u003C\u002Fstrong>. Limited training data results in less effective models, which discourages the use of Arabic language technology. This lack of adoption, in turn, reduces the incentive to invest in better Arabic language support, perpetuating the data shortage.\u003C\u002Fp>\u003Ch3 id=\"design-challenges\">Design challenges\u003C\u002Fh3>\u003Cp>Arabic's right-to-left writing system requires more than just text translation; it demands \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fblog\u002F6-challenges-of-localizing-your-app-to-arabic-and-how-to-solve-them#redesigning-the-ui-and-layout\">\u003Cstrong>a complete redesign of user interfaces\u003C\u002Fstrong>\u003C\u002Fa>. This includes mirroring elements like menus, buttons, text, and images. While LLMs don't manage these layout changes, their output needs to fit well into these reversed interfaces and adapt to specific cultural preferences.\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card kg-card-hascaption\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"1131\" height=\"846\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002Fimage.png 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F03\u002Fimage.png 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage.png 1131w\" sizes=\"(min-width: 720px) 720px\">\u003Cfigcaption>Example:\u003Ca href=\"https:\u002F\u002Fwww.aljazeera.net\u002Ftravel\u002F2025\u002F3\u002F15\u002F%D9%88%D8%AC%D9%87%D8%A9-%D8%B9%D8%B4%D8%A7%D9%82-%D8%A7%D9%84%D8%B3%D9%8A%D8%A7%D8%B1%D8%A7%D8%AA-%D8%B3%D9%8A%D8%A7%D8%AD%D8%A9-%D8%A7%D9%84%D8%B3%D8%B1%D8%B9%D8%A9-%D8%B9%D9%84%D9%89\"> an article on \u003C\u002Fa>Al Jazeera.\u003C\u002Ffigcaption>\u003C\u002Ffigure>\u003Cp>Visual elements like colors and images must be carefully chosen to resonate with the target audience. Achieving this requires tight cooperation between language experts, designers, and developers, significantly increasing the complexity of localization.\u003C\u002Fp>\u003Ch2 id=\"%F0%9F%A4%96-testing-8-llms-on-arabic-translation\">🤖 Testing 8 LLMs on Arabic translation\u003C\u002Fh2>\u003Cp>The Arabic language AI technology is promising, but it is developing slowly. The attempts to push further its emerging potential are faced with new ongoing hurdles. \u003Cstrong>These systems have certainly advanced, yet they remain works in progress\u003C\u002Fstrong> rather than final answers to Arabic natural language processing needs — especially when localizing content for Arabic-speaking users.\u003C\u002Fp>\u003Cp>Current Arabic-capable models demonstrate varied capabilities: some excel at formal text generation while stumbling with dialectal variations; others handle basic translation effectively but falter when it comes to cultural nuances. \u003C\u002Fp>\u003Cp>\u003Cstrong>The technology shows promise in structured contexts \u003C\u002Fstrong>like information retrieval and straightforward content creation, but requires human oversight for tasks demanding cultural sensitivity or technical precision.  Here's a summary of some Arabic LLMs' current state, capabilities, strengths, and weaknesses.\u003C\u002Fp>\u003Ch3 id=\"1-gpt-4-multilingual-model\">1. GPT-4 (multilingual model)\u003C\u002Fh3>\u003Cp>GPT-4 works across multiple languages, including Arabic, with the ability to process and produce text in over 25 languages. The model demonstrates proficiency in generating well-structured sentences in Modern Standard Arabic (MSA) while offering \u003Cstrong>some capability to handle various Arabic dialects\u003C\u002Fstrong>. \u003C\u002Fp>\u003Cp>It serves effectively for conversational tasks, making it \u003Cstrong>suitable for customer support, casual interactions, and information retrieval\u003C\u002Fstrong> in Arabic. Also, it provides reasonable translation services between Arabic and other languages, particularly when working with MSA rather than dialectal variants.\u003C\u002Fp>\u003Cp>The following prompt asks for the translation of \u003Cstrong>يا مساء الزبادي,\u003C\u002Fstrong> an informal way to say \"good evening\" in Egyptian Arabic. The translation is literal and doesn’t convey the playfulness of the greeting:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Flh7-rt.googleusercontent.com\u002Fdocsz\u002FAD_4nXdUr8hJfTLxcNe6Yf374TQyepcj-KNc5RgEZnA7U7KqzZnomdN-vEDZvWsc_wdTk6KU7Q1HNI8qJv-GiCS5ZRZ5Rd1_O6NiDTWUh6v90_Rrtak42M7ZpzdbKKElyjkggdI3HYeJ0YASwhUlp1QL6qM?key=CY7r7ITGL9faXpAb1Za7f00j\" class=\"kg-image\" alt loading=\"lazy\">\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Demonstrates a good general understanding of MSA and some Arabic dialects.\u003C\u002Fli>\u003Cli>Capable of generating creative content in Arabic, such as short stories, articles, and poetry.\u003C\u002Fli>\u003Cli>Maintains contextual coherence in responses, correctly using grammatical elements like tense and subject-object agreement.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Has difficulty with dialectal Arabic. For example, it might misinterpret an Egyptian Arabic phrase or translate it into MSA, losing the intended regional meaning.\u003C\u002Fli>\u003Cli>Struggles with informal, slang-filled Arabic speech.\u003C\u002Fli>\u003Cli>May misinterpret or create ambiguous words due to the common omission of vowels in written Arabic, which can lead to multiple meanings.\u003C\u002Fli>\u003C\u002Ful>\u003Cp>\u003C\u002Fp>\u003Ch3 id=\"2-claude\">2. Claude\u003C\u002Fh3>\u003Cp>Claude understands Modern Standard Arabic (MSA) texts accurately and handles complex Arabic grammar correctly. It processes  sentence structures, word forms, and language features properly. \u003Cstrong>The model works consistently well with different types of Arabic text\u003C\u002Fstrong>, including those with challenging grammar elements like dual forms and verb patterns.\u003C\u002Fp>\u003Cp>Here is an example in which Claude explains the translation and the meaning of the sentence:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage-1.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"778\" height=\"241\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002Fimage-1.png 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage-1.png 778w\" sizes=\"(min-width: 720px) 720px\">\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>It performs well in academic and formal settings.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Sometimes it misses cultural nuances.\u003C\u002Fli>\u003Cli>Struggles with dialect-specific expressions.\u003C\u002Fli>\u003Cli>Might produce unnatural-sounding responses in casual conversations.\u003C\u002Fli>\u003Cli>Shows inconsistency in technical terminology, potentially mixing Arabic terms and English loanwords.\u003C\u002Fli>\u003C\u002Ful>\u003Cp>Apart from global, Western models like ChatGPT and Claude, \u003Cstrong>there are several emerging models that focus exclusively on Arabic localization\u003C\u002Fstrong>. Some of these AI tools, designed with cultural and linguistic sensitivity, can leverage 'pronoia' to cultivate positive communication. This means that translations and content are accurate and intentionally crafted to resonate with users on a deeper level. Let's see how they fared in our analysis next.\u003C\u002Fp>\u003Ch3 id=\"3-jais\">3. Jais\u003C\u002Fh3>\u003Cp>\u003Ca href=\"https:\u002F\u002Finceptionai.ai\u002Fjais\u002Findex.html\">Jais\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fmbzuai.ac.ae\u002Fnews\u002Fmeet-jais-the-worlds-most-advanced-arabic-large-language-model-open-sourced-by-g42s-inception\u002F\">released by Inception in 2023,\u003C\u002Fa> was trained on 116 billion Arabic tokens and 279 billion English data tokens. \u003Cstrong>It excels at text generation, creating natural-sounding articles, reports, and stories in Arabic.\u003C\u002Fstrong> It is capable of generating natural Arabic text, sentiment analysis, summarization, \u003Ca href=\"https:\u002F\u002Fwww.techtarget.com\u002Fwhatis\u002Fdefinition\u002Fnamed-entity-recognition-NER\">Named Entity Recognition\u003C\u002Fa>, and Arabic-English translation. Also, it performs well in machine translation between Arabic and other languages.\u003C\u002Fp>\u003Cp>Let's use the same example of مساء الزبادي that we used for ChatGPT. We see that it fails, as it doesn't convey the playfulness of the expression. It says that the translation is literal and needs more context:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage--1--1.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"478\" height=\"389\">\u003C\u002Ffigure>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage--2-.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"563\" height=\"383\">\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>It was trained explicitly on Arabic data, providing a deep understanding of MSA and regional dialects.\u003C\u002Fli>\u003Cli>Highly effective in tasks like text summarization and sentiment analysis for formal Arabic.\u003C\u002Fli>\u003Cli>Designed for business integration: it offers document automation and customer service (chatbots).\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Still struggles with understanding and generating content in various regional Arabic dialects.\u003C\u002Fli>\u003Cli>Tends to produce formal MSA, limiting its usefulness for informal applications.\u003C\u002Fli>\u003Cli>May generate content lacking natural flow or authenticity in creative tasks.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"4-tarjamas-arabic-machine-translation-amt\">4. Tarjama's Arabic Machine Translation (AMT)\u003C\u002Fh3>\u003Cp>\u003Ca href=\"https:\u002F\u002Ftarjama.com\u002Famt\u002F\">Tarjama's Arabic Machine Translation (AMT)\u003C\u002Fa> system offers \u003Cstrong>business-focused translations with a focus on accuracy and security\u003C\u002Fstrong>. One of its strengths is that it maintains original document formatting and complies with \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fdictionary\u002FISO-27001\">ISO 27001\u003C\u002Fa> standards. AMT can be customized using client-specific data for enhanced precision and uses advanced technology for Arabic translation.\u003C\u002Fp>\u003Cp>The following example is about a proverb coined to indicate the extent of similarity and identity between a mother and her daughter. Regardless of getting a literal translation, it couldn’t differentiate between two Arabic words without diacritics:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Flh7-rt.googleusercontent.com\u002Fdocsz\u002FAD_4nXdv6VrZQMAmJpQ5lTM2TWXtb1agLDmZZ2G95fq2oGMJzndgQzhw1xQbO-lzbr7kcBi0o9uN9gpgskSSAKqTsuJ1RpO413dH-fRc4-N89mdOWLsgnE4tYXOYV7D7IYWi8U-a98ftJu6QU0P0pKm-fA?key=CY7r7ITGL9faXpAb1Za7f00j\" class=\"kg-image\" alt loading=\"lazy\">\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Trained on curated, high-quality data, which contributes to its accuracy.\u003C\u002Fli>\u003Cli>Handles domain-specific terminology (legal, financial, healthcare, etc.).\u003C\u002Fli>\u003Cli>Integrates into existing workflows, supporting various document formats.\u003C\u002Fli>\u003Cli>Incorporates a Human-in-the-Loop approach.\u003C\u002Fli>\u003Cli>Adheres to ISO 27001 standards.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Highly specialized jargon may require human review.\u003C\u002Fli>\u003Cli>Some idiomatic or culturally specific references may need post-editing.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"5-alibaba-clouds-qwen-series-qwen-15-qwen-2\">5. Alibaba Cloud's Qwen Series (Qwen 1.5 & Qwen 2)\u003C\u002Fh3>\u003Cp>Alibaba Cloud's \u003Ca href=\"https:\u002F\u002Fhuggingface.co\u002FQwen\">Qwen series\u003C\u002Fa>, including versions 1.5 and 2, supports over 29 languages including Arabic. \u003Cstrong>It performs well in transcribing and translating, improving speech translation\u003C\u002Fstrong>.\u003C\u002Fp>\u003Cp>Here is a translation example from Qwen where we can see it struggled with context and cultural accuracy:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage--3-.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"1143\" height=\"401\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002Fimage--3-.png 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F03\u002Fimage--3-.png 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002Fimage--3-.png 1143w\" sizes=\"(min-width: 720px) 720px\">\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Supports 29 languages, including Arabic.\u003C\u002Fli>\u003Cli>Offers a cost-effective alternative to human translation. \u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>May struggle with cultural nuances and idiomatic expressions.\u003C\u002Fli>\u003Cli>May encounter challenges with specialized or technical jargon.\u003C\u002Fli>\u003Cli>Can face difficulties in comprehending larger or implicit contexts.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"6-arabert\">6. AraBERT\u003C\u002Fh3>\u003Cp>\u003Ca href=\"https:\u002F\u002Fhuggingface.co\u002Faubmindlab\u002Fbert-base-arabert\">AraBERT \u003C\u002Fa> was \u003Cstrong>specifically designed for the Arabic language based on Google's BERT architecture\u003C\u002Fstrong>. It does a great job in tasks like Named Entity Recognition (NER), part-of-speech tagging, sentiment analysis, and text classification.\u003C\u002Fp>\u003Cp>Now, let's test this model with this \u003Cstrong>source text\u003C\u002Fstrong>: \u003C\u002Fp>\u003Cp>\u003Cem>\"This product will revolutionize the industry.\"\u003C\u002Fem>\u003C\u002Fp>\u003Cp>\u003Cstrong>AraBERT’s Target\u003C\u002Fstrong>: \"سوف تحدث هذه المنتجات ثورة في الصناعة.\"\u003C\u002Fp>\u003Cp>Here, we run into an incorrect word choice or mistranslation. The phrase \"سوف تحدث هذه المنتجات ثورة\" translates to \"these products will revolutionize.\" However, the original source talks about a singular \"product,\" so the plural form (\"المنتجات\") would be a meaning error. \u003C\u002Fp>\u003Cp>The correct translation should be \"سوف تحدث هذه المنتج ثورة في الصناعة.\"\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Efficient for formal MSA.\u003C\u002Fli>\u003Cli>Performs well on document classification, summarization, and question answering.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Poor performance with Arabic dialects.\u003C\u002Fli>\u003Cli>Weak in creative text generation and complex conversations.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"7-camel-camel-lab%E2%80%99s-arabic-models\">7. CAMeL (CAMeL Lab’s Arabic Models)\u003C\u002Fh3>\u003Cp>\u003Ca href=\"https:\u002F\u002Fhuggingface.co\u002FCAMeL-Lab\u002Fbert-base-arabic-camelbert-da\">CAMeL Lab's Arabic models\u003C\u002Fa> \u003Cstrong>specialize in various dialects: Gulf, Levantine, Egyptian, and Maghrebi\u003C\u002Fstrong>. Their suite of tools, known as CAMeL Tools, offers functionalities like pre-processing, morphological modeling, dialect identification, named entity recognition, and sentiment analysis.\u003C\u002Fp>\u003Cp>What about a translation example? Let's have a look at it. \u003C\u002Fp>\u003Cp>We will use this \u003Cstrong>source text\u003C\u002Fstrong>: \"The match was a real game-changer.\"\u003C\u002Fp>\u003Cp>\u003Cstrong>CAMel’s target translation\u003C\u002Fstrong>: \"كانت المباراة مغيرة حقيقية للعبة.\"\u003C\u002Fp>\u003Cp>We run into a literal translation of an idiomatic expression. The phrase \"game-changer\" is a colloquial expression in English, meaning something that changes the course of an event or situation. CAMeL could have translated this literally as \"مغيرة حقيقية للعبة,\" which would not convey the idiomatic meaning correctly.\u003C\u002Fp>\u003Cp>A better and more accurate translation here would be: \"كانت المباراة نقطة تحول حقيقية.\"\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Fine-tuned for dialectal Arabic.\u003C\u002Fli>\u003Cli>Handles sentiment analysis and code-switching.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Difficulties with contextual coherence in longer texts, especially mixed MSA and dialects.\u003C\u002Fli>\u003Cli>Struggles with highly informal or niche contexts due to being trained on limited data.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"8-tashkeela-qcri\">8. Tashkeela (QCRI)\u003C\u002Fh3>\u003Cp>Qatar Computing Research Institute's (QCRI) \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FAnwarvic\u002FArabic-Tashkeela-Model\">Tashkeela AI model\u003C\u002Fa> \u003Cstrong>adds diacritical marks to Arabic text and performs Named Entity Recognition (NER)\u003C\u002Fstrong> and other Natural Language Processing tasks in MSA. It features a corpus of 75 million fully vocalized words from 97 classical and modern Arabic books, which made possible the development of diacritization systems.\u003C\u002Fp>\u003Cp>Here is an example of using Tashkeela to add diacritization to Arabic texts to make them clearer:\u003C\u002Fp>\u003Cul>\u003Cli>\u003Cstrong>Arabic text without diacritics\u003C\u002Fstrong>:\u003Cbr>\"ذهب الولد إلى المدرسة\"\u003C\u002Fli>\u003Cli>\u003Cstrong>Arabic text with diacritics\u003C\u002Fstrong>:\u003Cbr>\"ذَهَبَ اَلْوَلَدُ إِلَى اَلْمَدْرَسَةِ\"\u003C\u002Fli>\u003Cli>\u003Cstrong>English translation\u003C\u002Fstrong>: \u003Cbr>\"The boy went to school.\"\u003C\u002Fli>\u003C\u002Ful>\u003Cblockquote>\u003Cem>\u003Cstrong>Note:\u003C\u002Fstrong>\u003C\u002Fem> This model is used to add diacritics to clarify the meaning and provide an accurate translation. It \u003Cem>doesn't \u003C\u002Fem>provide translation. Its accuracy is continuously increasing, becoming more and more reliable.\u003C\u002Fblockquote>\u003Cfigure class=\"kg-card kg-image-card kg-card-hascaption\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F04\u002FExample-of-Tashkeela.png\" class=\"kg-image\" alt loading=\"lazy\" width=\"1164\" height=\"281\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F04\u002FExample-of-Tashkeela.png 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F04\u002FExample-of-Tashkeela.png 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F04\u002FExample-of-Tashkeela.png 1164w\" sizes=\"(min-width: 720px) 720px\">\u003Cfigcaption>Tashkeela-Model (\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FAnwarvic\u002FArabic-Tashkeela-Model\">GitHub\u003C\u002Fa>)\u003C\u002Ffigcaption>\u003C\u002Ffigure>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-green\">\u003Cdiv class=\"kg-callout-emoji\">🏆\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Upsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>It reduces ambiguity through diacritics.\u003C\u002Fli>\u003Cli>With a focus on MSA, it performs well with standardized Arabic.\u003C\u002Fli>\u003C\u002Ful>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-red\">\u003Cdiv class=\"kg-callout-emoji\">😥\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Downsides\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cul>\u003Cli>Struggles with non-standard Arabic.\u003C\u002Fli>\u003Cli>Since it's limited to diacritics, it offers limited utility for deeper text understanding or generation.\u003C\u002Fli>\u003C\u002Ful>\u003Ch3 id=\"%E2%98%9D%EF%B8%8F-my-picks-as-a-translator\">☝️ My picks as a translator\u003C\u002Fh3>\u003Cp>Although all models are still a work in progress, \u003Cstrong>Claude\u003C\u002Fstrong> is my personal pick for Arabic localization tasks. The LLM by Anthropic is advancing constantly — it understands cultural references and considers the nuances of the language. It can also:\u003C\u002Fp>\u003Cul>\u003Cli>🚩 Detect literal translation and help revise long texts, suggesting better options when asked. \u003C\u002Fli>\u003Cli>✏️ Help with different types of content, providing accurate equivalents, and considering context.\u003C\u002Fli>\u003Cli>💬 Remind the user of the different word meanings that could be used depending on the context.\u003C\u002Fli>\u003C\u002Ful>\u003Cp>For simple translation, \u003Cstrong>ChatGPT \u003C\u002Fstrong>is also a good option, considering how it understands different dialects. It can help identify the nature of the source and suggest a suitable translation. However, it might skip some paragraphs or lines.\u003C\u002Fp>\u003Cp>Finally, \u003Cstrong>AMT\u003C\u002Fstrong> and \u003Cstrong>Qwen\u003C\u002Fstrong> are suitable options if you're dealing with business and customer service-related content, but keep in mind that they don't always get dialects correctly. They are progressing, but until now, I haven't found a better option than Claude.\u003C\u002Fp>\u003Cblockquote>🎙️ How realistic are agentic workflows in localization, and are we ready to implement them? \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fblog\u002Fmost-localization-teams-arent-ready-for-ai-workflows-bridging-the-gap-s02-ep08\">Listen to our Bridging the Gap podcast episode with Julia Díez\u003C\u002Fa> for a deep dive into it.\u003C\u002Fblockquote>\u003Ch2 id=\"%F0%9F%A4%B7-can-ai-deal-with-my-arabic-content\">🤷 Can AI deal with my Arabic content?\u003C\u002Fh2>\u003Cp>Now, you might ask, \u003Cstrong>\"How can I know if the LLM I use suits my content translation needs?\u003C\u002Fstrong>\". Well, that's a fair question. Here are some factors you should take into consideration to assess this:\u003C\u002Fp>\u003Cfigure class=\"kg-card kg-image-card\">\u003Cimg src=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F03\u002FAdobeStock_1102016803.jpeg\" class=\"kg-image\" alt loading=\"lazy\" width=\"2000\" height=\"1334\" srcset=\"https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw600\u002F2025\u002F03\u002FAdobeStock_1102016803.jpeg 600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1000\u002F2025\u002F03\u002FAdobeStock_1102016803.jpeg 1000w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw1600\u002F2025\u002F03\u002FAdobeStock_1102016803.jpeg 1600w, https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002Fsize\u002Fw2400\u002F2025\u002F03\u002FAdobeStock_1102016803.jpeg 2400w\" sizes=\"(min-width: 720px) 720px\">\u003C\u002Ffigure>\u003Ch3 id=\"1-industry-specificity\">1. Industry specificity\u003C\u002Fh3>\u003Cp>Check if the model is trained on domain-specific data (legal, medical, etc.) and if your content is technical\u002Fspecialized. You need to determine how standardized the terminology is.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Look for fine-tuned models (AraBERT, CAMeL) or industry benchmarks.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"2-target-audienceregion\">2. Target audience\u002Fregion\u003C\u002Fh3>\u003Cp>Ensure the model handles dialectal variations (Egyptian, Gulf, etc.). Check which Arabic variants are used and if there are multiple dialects needed or a required formality level.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Check dialect support (CAMeL) vs. MSA defaults.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"3-user-experience\">3. User experience\u003C\u002Fh3>\u003Cp>Verify contextual relevance and coherent dialogue in conversations.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Test conversational flows (GPT-4) and dialect-specific training.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"4-informalsocial-media-language\">4. Informal\u002Fsocial media language\u003C\u002Fh3>\u003Cp>If your content includes more informal language, check how the model handles slang, abbreviations, and code-switching.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Test with social media samples; note limitations of MSA-focused models.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"5-nuanced-sentiment-analysis\">5. Nuanced sentiment analysis\u003C\u002Fh3>\u003Cp>Test how the LLM handles complex emotions, like sarcasm and ambiguity.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Run tests with ambiguous\u002Femotional phrases; consider AraBERT or CAMeL.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"6-undiacritized-arabic\">6. Undiacritized Arabic\u003C\u002Fh3>\u003Cp>Test the model's ability to handle text without vowel markings.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Use Tashkeela for diacritization, and assess general LLM context inference.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"7-multilingualcode-switched-content\">7. Multilingual\u002Fcode-switched content\u003C\u002Fh3>\u003Cp>Check for fluid language switching and bilingual content processing.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Test with Arabic-English or other mixed language content.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"8-content-formatstyle-variability\">8. Content format\u002Fstyle variability\u003C\u002Fh3>\u003Cp>Ensure the model handles different registers and contexts.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Test with diverse text types (news, conversations) for contextual appropriateness.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"9-real-time-processing\">9. Real-time processing\u003C\u002Fh3>\u003Cp>For time-sensitive applications, evaluate processing times of the different options.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong> Test response time and processing speed, especially in real-time scenarios.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"10-accuracy-vs-functionality\">10. Accuracy vs. functionality\u003C\u002Fh3>\u003Cp>Determine the required accuracy level for your specific use case.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Check accuracy rates across various tasks (translation, summarization).\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"11-performance-evaluation\">11. Performance evaluation\u003C\u002Fh3>\u003Cp>Monitor overall tendencies like accuracy, dialect handling, and technical term consistency.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Use side-by-side human review to score translations on clarity and accuracy.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"12-error-analysis\">12. Error analysis\u003C\u002Fh3>\u003Cp>Finally, document dialect mismatches, technical term confusion, formality issues, and cultural misunderstandings. Evaluate error significance (safety, brand, user experience, legal). Decide how much human oversight is needed.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Track the top error types with color-coded severity ratings (red\u002Fyellow\u002Fgreen).\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Ch3 id=\"13-final-check-sample-data-testing\">13. Final check: sample data testing\u003C\u002Fh3>\u003Cp>Crucially, always test with real sample data to identify specific strengths and weaknesses. Create representative samples with dialects, technical terms, formality variations, numbers, and dates.\u003C\u002Fp>\u003Cdiv class=\"kg-card kg-callout-card kg-callout-card-yellow\">\u003Cdiv class=\"kg-callout-emoji\">👉\u003C\u002Fdiv>\u003Cdiv class=\"kg-callout-text\">\u003Cstrong>Practical tip: \u003C\u002Fstrong>Test with a mix of everyday and technical Arabic text samples.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cp>After considering all of these aspects, you can go ahead and use the AI model of your choice for a while to see how it performs. But if you're still unsure about using AI in your industry, this would be \u003Cstrong>a quick and brief categorization\u003C\u002Fstrong> that tells you if it would make sense to use AI in your industry or not: \u003C\u002Fp>\u003Cul>\u003Cli>\u003Cstrong>🟢 Safe for AI\u003C\u002Fstrong>:\u003Cstrong> \u003C\u002Fstrong>General MSA information, basic queries, non-critical content, and standard communication.\u003C\u002Fli>\u003Cli>\u003Cstrong>🟠 Requires careful monitoring\u003C\u002Fstrong>:\u003Cstrong> \u003C\u002Fstrong>Mixed dialects, semi-technical documentation, marketing, education.\u003C\u002Fli>\u003Cli>\u003Cstrong>🔴 Not ready for AI\u003C\u002Fstrong>:\u003Cstrong> \u003C\u002Fstrong>Legal documents, medical instructions, safety-critical information, creative content, and complex manuals.\u003C\u002Fli>\u003C\u002Ful>\u003Ch2 id=\"%F0%9F%91%80-conclusion-is-it-worth-it-to-use-ai-to-localize-to-arabic\">👀 Conclusion: Is it worth it to use AI to localize to Arabic?\u003C\u002Fh2>\u003Cp>\u003Cstrong>AI is a valuable assistant for Arabic content, but has limitations\u003C\u002Fstrong>. It still makes mistakes regardless of what model you choose, especially when the material is nuanced and contains specialized vocabulary. \u003C\u002Fp>\u003Cp>Arabic's complexity requires contextual understanding that AI alone cannot fully provide. While AI can translate and generate content to some point, \u003Cstrong>human oversight is what controls the cultural relevance, proper right-to-left layouts, and brand alignment\u003C\u002Fstrong>. Dialectal Arabic particularly challenges AI systems, often necessitating human adaptation. \u003C\u002Fp>\u003Cp>Critical documents like legal or medical texts absolutely require human review. If you're dealing with that type of sensitive material (or with technical or very nuanced content in Arabic), proofreading is still highly important. We're here to help — get in touch with expert Arabic translators from our \u003Ca href=\"https:\u002F\u002Flocalazy.com\u002Fdocs\u002Fgeneral\u002Fcontinuous-localization-team\">Continuous Localization Team\u003C\u002Fa>, or begin experimenting with Localazy on your own for your Arabic-speaking projects.\u003C\u002Fp>","public","none","2025-03-04T13:53:26.000+01:00",[1968],{"id":1855,"name":1856,"slug":1857,"profile_image":1858,"cover_image":10,"bio":1969,"website":10,"location":1970,"facebook":10,"twitter":10,"meta_title":10,"meta_description":10,"url":1971},"English-Arabic translator and localization specialist immersed in the world of transcreation and localization. Author of Egypt Localization Guide. ","Egypt","https:\u002F\u002Fghost.localazy.com\u002Fauthor\u002Feman-abdo\u002F",[1973,1978,1981,1984,1989,1994,1999,2004,2007,2012],{"id":1974,"name":1975,"slug":1976,"description":1878,"feature_image":10,"visibility":1964,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"codeinjection_head":10,"codeinjection_foot":10,"canonical_url":10,"accent_color":10,"url":1977},"66a25e3f34f1ce00011a1fb2","[DICT] Large Language Model (LLM)","term-185","https:\u002F\u002Fghost.localazy.com\u002Ftag\u002Fterm-185\u002F",{"id":1979,"name":1866,"slug":1867,"description":10,"feature_image":10,"visibility":1964,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"codeinjection_head":10,"codeinjection_foot":10,"canonical_url":10,"accent_color":10,"url":1980},"64ca4e64047091000193df73","https:\u002F\u002Fghost.localazy.com\u002Ftag\u002Fchatgpt\u002F",{"id":1982,"name":1830,"slug":1831,"description":10,"feature_image":10,"visibility":1964,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"codeinjection_head":10,"codeinjection_foot":10,"canonical_url":10,"accent_color":10,"url":1983},"64ca4e64047091000193df72","https:\u002F\u002Fghost.localazy.com\u002Ftag\u002Fai\u002F",{"id":1985,"name":1986,"slug":1987,"description":10,"feature_image":10,"visibility":1964,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"codeinjection_head":10,"codeinjection_foot":10,"canonical_url":10,"accent_color":10,"url":1988},"60fffb486f8ebe0001bd8ddd","[COMSHARE]","comshare","https:\u002F\u002Fghost.localazy.com\u002Ftag\u002Fcomshare\u002F",{"id":1990,"name":1991,"slug":1992,"description":1902,"feature_image":10,"visibility":1964,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"codeinjection_head":10,"codeinjection_foot":10,"canonical_url":10,"accent_color":10,"url":1993},"664b1d5334f1ce00011a04ac","[DICT] RTL","term-161","https:\u002F\u002Fghost.localazy.com\u002Ftag\u002Fterm-161\u002F",{"id":1995,"name":1996,"slug":1997,"description":1909,"feature_image":10,"visibility":1964,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"codeinjection_head":10,"codeinjection_foot":10,"canonical_url":10,"accent_color":10,"url":1998},"66abc1ee34f1ce00011a2606","[DICT] Neural Machine Translation (NMT)","term-189","https:\u002F\u002Fghost.localazy.com\u002Ftag\u002Fterm-189\u002F",{"id":2000,"name":2001,"slug":2002,"description":1958,"feature_image":10,"visibility":1964,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"codeinjection_head":10,"codeinjection_foot":10,"canonical_url":10,"accent_color":10,"url":2003},"636ba09f0470910001937216","[CTA] Order Professional Translations 💎","cta-31","https:\u002F\u002Fghost.localazy.com\u002Ftag\u002Fcta-31\u002F",{"id":2005,"name":1870,"slug":1871,"description":1870,"feature_image":10,"visibility":1964,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"codeinjection_head":10,"codeinjection_foot":10,"canonical_url":10,"accent_color":10,"url":2006},"60a37fc7cf7b6a0001d1f0d8","https:\u002F\u002Fghost.localazy.com\u002Ftag\u002Flocales\u002F",{"id":2008,"name":2009,"slug":2010,"description":1931,"feature_image":10,"visibility":1964,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"codeinjection_head":10,"codeinjection_foot":10,"canonical_url":10,"accent_color":10,"url":2011},"68110b1663350c0001d5985c","[DICT] Named Entity Recognition (NER)","term-294","https:\u002F\u002Fghost.localazy.com\u002Ftag\u002Fterm-294\u002F",{"id":2013,"name":2014,"slug":2015,"description":1943,"feature_image":10,"visibility":1964,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"codeinjection_head":10,"codeinjection_foot":10,"canonical_url":10,"accent_color":10,"url":2016},"6810f9a963350c0001d596ac","[DICT] Natural Language Processing (NLP)","term-293","https:\u002F\u002Fghost.localazy.com\u002Ftag\u002Fterm-293\u002F",{"id":1855,"name":1856,"slug":1857,"profile_image":1858,"cover_image":10,"bio":1969,"website":10,"location":1970,"facebook":10,"twitter":10,"meta_title":10,"meta_description":10,"url":1971},{"id":1974,"name":1975,"slug":1976,"description":1878,"feature_image":10,"visibility":1964,"og_image":10,"og_title":10,"og_description":10,"twitter_image":10,"twitter_title":10,"twitter_description":10,"meta_title":10,"meta_description":10,"codeinjection_head":10,"codeinjection_foot":10,"canonical_url":10,"accent_color":10,"url":1977},"https:\u002F\u002Fghost.localazy.com\u002Fai-8-llm-arabic-models-tested-to-translate\u002F",[2021,2037,2053],{"id":2022,"owner":2023,"created_by":10,"sort":10,"title":2024,"slug":2025,"modified_on":877,"created_on":2026,"pinned":160,"badge":10,"priority":1824,"main_image":2027,"status":5,"tags":2028,"dictionary":2036,"reading_time":1835,"excerpt":-1,"og_title":877,"og_description":877,"og_image":877},"6953a75663350c0001d6289c",{"id":1817,"first_name":1818,"last_name":877,"slug":1819,"avatar":1820},"A practical guide to translating with Localazy AI","a-practical-guide-to-translating-with-localazy-ai","2025-12-30T17:08:51.000+01:00","https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F12\u002FLocalazy-AI-in-practice.png",[2029,2030,2033],{"id":402,"created_on":1828,"status":5,"label":1830,"slug":1831,"on_index_page":162},{"id":451,"created_on":1828,"status":5,"label":2031,"slug":2032,"on_index_page":160},"Automated Localization","automated-localization",{"id":451,"created_on":1828,"status":5,"label":2034,"slug":2035,"on_index_page":160},"Guide","guide",[],{"id":2038,"owner":2039,"created_by":10,"sort":10,"title":2044,"slug":2045,"modified_on":877,"created_on":2046,"pinned":160,"badge":10,"priority":1824,"main_image":2047,"status":5,"tags":2048,"dictionary":2052,"reading_time":1835,"excerpt":-1,"og_title":877,"og_description":877,"og_image":877},"649de9ec047091000193d3ab",{"id":2040,"first_name":2041,"last_name":877,"slug":2042,"avatar":2043},"61dc4f36df4bcb00010bd510","Gletjan Muco","gletjan-muco","https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2025\u002F09\u002FProfile-Pic---Blue-Background.jpeg","Translation technology: Exploring the pros and cons of AI translation","translation-technology-exploring-the-pros-and-cons-of-ai-translation","2023-08-02T15:02:12.000+02:00","https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2023\u002F07\u002FAI-Translation_Featured.png",[2049,2050,2051],{"id":402,"created_on":1828,"status":5,"label":1830,"slug":1831,"on_index_page":162},{"id":402,"created_on":1828,"status":5,"label":1866,"slug":1867,"on_index_page":160},{"id":451,"created_on":1828,"status":5,"label":352,"slug":1833,"on_index_page":162},[],{"id":2054,"owner":2055,"created_by":10,"sort":10,"title":2060,"slug":2061,"modified_on":877,"created_on":2062,"pinned":160,"badge":10,"priority":1824,"main_image":2063,"status":5,"tags":2064,"dictionary":2070,"reading_time":1835,"excerpt":-1,"og_title":877,"og_description":877,"og_image":877},"668fcad334f1ce00011a1c3a",{"id":2056,"first_name":2057,"last_name":877,"slug":2058,"avatar":2059},"65647a510470910001942876","Dorota Pawlak","dorota-pawlak","https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2024\u002F07\u002Fdorota.png","How to use ChatGPT to your advantage in translation (with examples)","how-to-use-chatgpt-to-your-advantage-in-translation-with-examples-gpt-series-1","2024-07-22T11:51:27.000+02:00","https:\u002F\u002Fghost.localazy.com\u002Fcontent\u002Fimages\u002F2024\u002F07\u002Flocalazy---AI-series.png",[2065,2066,2067],{"id":402,"created_on":1828,"status":5,"label":1830,"slug":1831,"on_index_page":162},{"id":402,"created_on":1828,"status":5,"label":1866,"slug":1867,"on_index_page":160},{"id":451,"created_on":1828,"status":5,"label":2068,"slug":2069,"on_index_page":160},"Translations","translations",[],{"id":4,"sort":10,"created_on":2072,"name":2073,"cta_link":2074,"banner_image":2075},"2022-03-17T12:23:14.000Z","default","https:\u002F\u002Flocalazy.com\u002Ffeatures\u002Flocalazy-ai",{"id":2076},"72defd59-a21d-44b9-822c-4e183ca62b99"]