Qwen from Alibaba has surpassed the 3 billion download mark: Chinese open-source models are capturing the market
Chinese corporation Alibaba has reached an impressive milestone: over the past six months, downloads of its open-source AI models from the Qwen family have exceeded 3 billion. This is a landmark event that demonstrates a shift in the global balance in the field of open artificial intelligence.
To date, Alibaba has released more than 460 neural networks into the public domain, which have served as the basis for the creation of 300,000 derivative models by developers worldwide. These figures underscore the scale of the ecosystem built around Qwen.
Hugging Face Data: Unquestionable Leadership
An analysis of the Hugging Face platform, published in the semi-annual report on the state of open models, paints an even more detailed picture. Over seven months of the current year, Qwen was downloaded 2.05 billion times, and the number of derivative repositories reached 151,448. For comparison, Google's figure stands at only 82,506.
The gap with other giants is also impressive: Google model downloads in 2026 are estimated at 418 million, and Meta's at 227 million. It is important to note that Hugging Face statistics only account for activity within the platform's ecosystem, excluding API requests and private deployments, making the real gap even more significant.
Experts emphasize that these figures should not be perceived as a direct indicator of market share, but they are an accurate barometer of popularity and community trust. The growth dynamics are impressive: the number of Qwen derivative repositories increases by 180-210 units daily. Notably, out of 28,531 GGUF conversions of these models, only 54 were created by Alibaba itself — the rest is the work of the community.
The Secret of Qwen's Success
Three key factors explain Qwen's dominance: regular updates, coverage of all model scales — from sub-billion versions to the giant Qwen3.8-Max with 2.4 trillion parameters — and the Apache 2.0 license, which removes all restrictions on commercial use and modification.
The decisive factor was the breadth of the model lineup. Platform data shows that versions with fewer than 1 billion parameters account for 83% of all downloads, while giant neural networks exceeding 100 billion account for only 1%. This explains why laboratories focused on large LLMs are lagging behind: Moonshot AI, which releases almost no models smaller than 70 billion, has gathered only 37 million downloads — 55 times fewer than Qwen.
The situation is similar in the local deployment segment: Qwen GGUF builds are downloaded 39.6 million times per month, Gemma's figure is 20.8 million, and Llama's is only 7.5 million.
The Geopolitics of Open AI
The report reveals a fundamental shift. In almost every month of the current year, the largest open model from China surpassed all American releases in size. The ceiling for Chinese models ranged from 754 billion to 2.78 trillion parameters, while for US competitors it did not exceed 130 billion in five out of seven months.
Licensing policies also differ dramatically. In China, 59% of models with more than 20 billion parameters are released under Apache 2.0, and 22% under MIT. None of them have restrictions on commercial use. In the US, the picture is the opposite: only 29% under open licenses, 41% under proprietary terms, and 30% with no license specified at all.
It is also telling that in the US, the leaders in the number of new open models were not AI laboratories but chip manufacturers: AMD and Nvidia each released more than 200 repositories, leaving Google and Meta behind.
My analysis: Qwen's success is not just download statistics, but a marker of the redistribution of influence in the global AI industry. China has bet on openness and accessibility, effectively turning its models into the de facto standard for developers. While the US tries to maintain control through infrastructure, China is winning over the minds and wallets of developers worldwide, creating a sustainable ecosystem that will be extremely difficult to displace.