Alibaba unveils mega AI model to take on OpenAI and DeepMind

Alibaba has released its most powerful artificial intelligence model to date, Qwen-3-Max-Preview, marking a major leap forward in the company’s AI ambitions.

The model is the first from Alibaba to surpass one trillion parameters, putting it in direct competition with the most advanced systems developed by OpenAI and Google DeepMind.

The model is being made available through Alibaba Cloud as well as select developer platforms, though it remains closed-source for now. Developers and businesses can access it via paid APIs and partner integrations, including its adoption as the default AI in certain coding assistants.

In terms of performance, Qwen-3-Max-Preview significantly outpaces Alibaba’s previous flagship models and delivers results on par with some of the most capable AI systems currently on the market, such as Anthropic’s Claude, DeepSeek, and Kimi. It boasts a massive context window of 262,000 tokens, allowing it to process and generate exceptionally long conversations and documents, supported by context caching technology for greater efficiency.

Pricing for the model follows a tiered, token-based system, designed to make it affordable for shorter tasks but scaling up in cost for large, complex workloads. This positions the model as both a research tool and a practical solution for enterprise applications ranging from natural language processing to software development support.

The launch highlights Alibaba’s determination to remain a serious player in the global AI race. At a time when some rivals are emphasizing smaller, more efficient models, Alibaba is signaling confidence in the power of scale, betting that trillion-parameter systems will define the next generation of breakthroughs in artificial intelligence.

Copied