China’s Ministry of Industry and Information Technology (MIIT) reports that its top three telecom providers—China Mobile, China Telecom, and China Unicom—have fully adopted the open-source DeepSeek AI model.
These companies use the DeepSeek-R1 model in various scenarios and products, offering tailored computing power solutions and supportive environments.
AI continues to drive innovation in telecom services. During the 2025 Spring Festival, telecom operators combined AI with 5G networks, cloud platforms, and big data to broaden applications and improve AI-driven services.
They enhanced digital tools using AI and 5G by delivering billions of precise location services for smart vehicle navigation over the holidays. According to MIIT, they also introduced AI-based voice inspection services to help financial institutions detect fraud and developed cloud computing tools to support efficient remote work.
AI and big data were further used to assist government operations. China Unicon monitored Spring Festival travel patterns and tourist hotspots in real-time, analysed key infrastructure, and tracked holiday spending trends and post-holiday economic recovery.
Alibaba Releases Qwen 2.5 to Compete with DeepSeek-V3
Separately, Alibaba unveiled a new version of its AI model, Qwen 2.5, which it claims surpasses the widely recognized DeepSeek-V3. The Qwen 2.5-Max model was launched on Chinese New Year, alongside the release of its visual model, Qwen 2.5-VL, which was made open source days later.
Alibaba also provided cloud computing support for the live broadcast of the Spring Festival Gala, a variety show watched by billions. The show featured music, dance, opera, martial arts, and comedy.
“Qwen 2.5-Max performs better overall compared to GPT-4o, DeepSeek-V3, and Llama-3.1-405B,” Alibaba Cloud stated on its official WeChat account, referencing OpenAI and Meta’s latest open-source AI models.
The release of DeepSeek’s AI assistant on January 10, powered by the DeepSeek-V3 model, along with the launch of its R1 model, has sent shockwaves through Silicon Valley. The Chinese startup’s reportedly low development and usage costs have raised questions about the high spending by leading U.S. AI companies.
DeepSeek’s recent achievements have intensified competition among domestic rivals to improve their AI technologies.
Two days after the DeepSeek-R1 launch, ByteDance, the parent company of TikTok, updated its flagship AI model. The company claimed this model outperformed Microsoft-backed OpenAI’s o1 in AIME, a benchmark test for understanding and responding to complex instructions.
This echoed DeepSeek’s assertion that its R1 model could compete with OpenAI’s o1 on several performance metrics.
