Alibaba Unveils Qwen 2.5-Max AI Model Amid Fierce Domestic Competition

BEIJING, China — Alibaba, the Chinese tech giant, unveiled its latest artificial intelligence model, Qwen 2.5-Max, on Wednesday, claiming it outperforms leading global AI systems, including OpenAI‘s GPT-4o and Meta‘s Llama-3.1-405B. The release coincided with the Lunar New Year holiday, a time when most Chinese workers are on break, highlighting the intense pressure Alibaba faces from domestic competitors like DeepSeek and ByteDance.

In a statement posted on its official WeChat account, Alibaba’s cloud unit asserted that Qwen 2.5-Max surpasses its rivals “almost across the board.” The announcement comes just weeks after DeepSeek, a Chinese AI startup, released its DeepSeek-V3 and R1 models, which have disrupted the global AI landscape and prompted a reevaluation of spending by U.S. tech firms.

DeepSeek’s rapid rise has forced Chinese tech companies to accelerate their AI development. ByteDance, the parent company of TikTok, recently launched its own AI model, which it claims outperforms OpenAI’s o1 in the AIME benchmark test, a measure of AI’s ability to handle complex instructions. Similarly, DeepSeek’s R1 model has been touted as a rival to OpenAI’s o1 in multiple performance benchmarks.

The competition intensified last May when DeepSeek released its open-source DeepSeek-V2 model, priced at just 1 yuan ($0.14) per 1 million tokens. This unprecedented affordability led to a 97% price reduction across Alibaba’s cloud services and prompted other Chinese tech giants, including Baidu and Tencent, to follow suit.

Liang Wenfeng, a key figure at DeepSeek, emphasized in a July interview with Chinese media outlet Waves that the startup prioritizes innovation over price wars. “We did not care about price wars,” Liang said, adding that achieving artificial general intelligence (AGI) — systems that surpass human capabilities in economically valuable tasks — remains DeepSeek’s primary goal.

Unlike Alibaba and other tech giants, DeepSeek operates as a lean research lab, staffed primarily by young graduates and doctoral students from top Chinese universities. Liang contrasted this approach with the high costs and rigid structures of larger companies, suggesting that the future of AI innovation may favor smaller, more agile organizations.

As the AI race heats up, Alibaba’s latest release underscores the growing competition within China’s tech sector, where startups like DeepSeek are challenging established players and reshaping the global AI landscape.