{"id":12011,"date":"2025-08-20T14:56:43","date_gmt":"2025-08-20T14:56:43","guid":{"rendered":"https:\/\/www.gsma.com\/get-involved\/gsma-foundry\/?p=12011"},"modified":"2026-04-22T18:21:46","modified_gmt":"2026-04-22T18:21:46","slug":"bridging-the-language-gap-how-kaz-llm-is-transforming-ai-for-kazakhstan-and-beyond","status":"publish","type":"post","link":"https:\/\/www.gsma.com\/get-involved\/gsma-foundry\/artificial-intelligence\/bridging-the-language-gap-how-kaz-llm-is-transforming-ai-for-kazakhstan-and-beyond\/","title":{"rendered":"Bridging the Language Gap: How KAZ LLM is Transforming AI for Kazakhstan and Beyond"},"content":{"rendered":"\n

In an era where Large Language Models (LLMs) are reshaping industries, a quiet revolution is unfolding, not in Silicon Valley, but in Kazakhstan.<\/p>\n\n\n\n

Oleksii Sharavar from QazCode, recently joined the GSMA Foundry team to share the story of KAZ LLM; an ambitious project to develop a Kazakh-language AI model tailored for local communities, businesses, and public services. What started as a small internal initiative has now become a model of digital inclusion and international collaboration.<\/p>\n\n\n\n

\n
\n
\"A<\/figure>\n<\/div>\n\n\n\n
<\/div>\n<\/div>\n\n\n\n
<\/div>\n\n\n\n

Why KAZ LLM? A Mission Born from a Language Gap<\/strong><\/p>\n\n\n\n

Kazakhstan\u2019s tech landscape is evolving rapidly, but one major challenge stood in the way: mainstream LLMs like ChatGPT lacked support for the Kazakh language.<\/p>\n\n\n\n

\u201cWe had over a million monthly requests from customers, and we couldn\u2019t respond to them fluently in Kazakh. There was no LLM support for our language. That\u2019s when we knew we had to act,\u201d Oleksii shared.<\/p>\n\n\n\n

Originally part of Beeline Kazakhstan\u2019s IT team, the QazCode unit realised that local language communities were being left behind by global AI tools. And it wasn\u2019t just Kazakh. This was a challenge shared across many under-represented languages.<\/p>\n\n\n\n

<\/div>\n\n\n\n

From an Internal Tool to a National Asset<\/strong><\/p>\n\n\n\n

Three years ago, the team began collecting voice data from call centers, transcribing and polishing it to build the first dataset of its kind. The result? Kaz-RoBERTA, a 2-billion-token Kazakh language model published on Hugging Face.<\/p>\n\n\n\n

\u201cWe forgot about it after publishing. But then we checked back, 3,000 downloads. That was our turning point. Clearly, others needed this too.\u201d<\/p>\n\n\n\n

This discovery spurred a broader vision: a large-scale, national AI project that would require deep collaboration. That\u2019s when partnerships with the Digital Ministry of Kazakhstan, Nazarbayev University, and the Barcelona Supercomputing Center came into play.<\/p>\n\n\n\n

<\/div>\n\n\n\n

Building a KAZ LLM: A Global Effort with Local Roots<\/strong><\/p>\n\n\n\n

With a clear goal and the right partners, KAZ LLM took shape. Over nine months, the team collected massive datasets, tackled technical roadblocks, and trained a 70-billion parameter multilingual model in Kazakh, Russian, Turkish, and English.<\/p>\n\n\n\n

\u201cIn countries like Kazakhstan, people often switch languages mid-sentence. Our model needed to understand that kind of linguistic fluidity,\u201d Oleksii noted.<\/p>\n\n\n\n

Access to academic literature, not just web forums or user-generated content, was crucial. The team worked closely with universities and government bodies to integrate high-quality, local data.<\/p>\n\n\n\n

<\/div>\n\n\n\n

AI with Impact: Education, SMEs, and Public Services<\/strong><\/p>\n\n\n\n

KAZ LLM\u2019s first real-world use QazCodee launched under the name AI Tutor within the Janymda app, an all-in-one lifestyle platform used by millions in Kazakhstan.<\/p>\n\n\n\n

For Students:<\/strong><\/p>\n\n\n\n