Russian businesses are demonstrating an unprecedented appetite for large language models (LLMs). In the first half of 2026, token consumption on the Yandex AI Studio platform reached 597 billion—18 times more than in the same period last year. Moreover, more than half of this volume came from Yandex's own developments.

The key trend is the rapid strengthening of Chinese vendors' positions. While Yandex models accounted for 54% of all consumption, Qwen (21.2%) and DeepSeek (13.5%) confidently took second and third places. American GPT-OSS from OpenAI has to settle for just 7.5%, indicating a radical shift in preferences on the Russian market.

Chinese models are squeezing out competitors

The dynamics on other cloud platforms only confirm this trend. On MWS Cloud, token consumption for the half-year grew 11 times compared to all of 2025, reaching 400 billion. Here, the undisputed leader is Qwen with a result of 261.1 billion tokens. GLM (91.2 billion) is in second place, and Kimi (81.4 billion) took third, pushing DeepSeek down. A similar picture was recorded on Cloud.ru, where demand grew 18 times over eight months, with a significant share of requests for models from China.

This sharp jump is not just statistics. Businesses have moved from point experiments to industrial use of generative technologies. Companies are selecting models for specific tasks, and foreign developments are becoming an integral part of production processes.

How much is the Russian LLM market worth

According to my estimates, based on data from market participants, the total volume of the LLM market in Russia in 2026 will grow by 35% and amount to about 19.6 billion rubles. The cloud segment, including subscriptions and API access, will reach almost 1.5 billion rubles. However, this figure is disputed: at Turbo Cloud (owned by Rostelecom), the forecast is considered underestimated. There, the real commercial capacity of LLM access by the end of the year is estimated at least at 10–12 billion rubles.

This dispute fits into a more global picture. AIANA analysts calculated that the entire AI market in the country grew by 29.7% in 2025, reaching 316.1 billion rubles, and by 2030 the baseline scenario suggests growth to 830 billion. A significant portion of these funds is now being directed by businesses not to purchasing equipment, but to renting capacity and ready-made solutions.

It is telling that 49% of Russian employers already use AI in hiring, and the largest banks earn tens and hundreds of billions of rubles from neural networks. Sber alone estimated the effect of AI for 2025 at more than 475 billion. The growing token consumption on cloud platforms is a direct reflection of this demand.

My conclusion: Chinese LLMs have become not just an alternative, but the de facto standard for Russian businesses. This is an irreversible structural shift that will only deepen, especially given restrictions on Western technologies. Investors and developers should closely monitor this trend—it determines not only the AI market, but also the future of the country's entire digital economy.