Reuters reported that on January 29, technology company Alibaba (China) announced the launch of a new version of the artificial intelligence (AI) model Qwen 2.5 with capabilities that the manufacturer asserted are superior to the AI model that is currently taking the market by storm, DeepSeek V-3.
In an announcement on its official WeChat account, Alibaba claimed: “Qwen 2.5-Max outperforms GPT-4o, DeepSeek-V3, and Llama-3.1-405B in almost every way.”
Also according to Reuters, the launch of Qwen 2.5-Max on the first day of the Lunar New Year in China shows that DeepSeek's strong rise in the past 3 weeks has not only put pressure on foreign competitors but also weighed down on domestic competitors.
In recent days, the startup DeepSeek has been continuously releasing upgraded versions of its AI model that operates without much cost or energy. Among them, DeepSeek-V3 was released on January 10, followed by DeepSeek-R1 released on January 20.
This sent shockwaves around the world, especially in the US, sending tech stocks plunging. The emergence of DeepSeek-R1 has made investors question the massive spending plans of leading AI companies in the US.
In the domestic market, not only Alibaba but also Bytedance (TikTok's parent company) has joined the AI race in China. Just 2 days after the launch of DeepSeek-R1, Bytedance immediately released an upgrade to its AI model, claiming that it outperformed OpenAI's GPT-o1 after comparing and evaluating its capabilities based on the AIME test - a standard for evaluating AI's ability to understand and respond to complex instructions.
