By Selena Li and Brenda Goh
DAVOS/SHANGHAI, Jan 23 (Reuters) - Lenovo is seeking partnerships with multiple large language models around the world to power its devices as it aims to establish itself as a global AI player, the company's chief financial officer said.
Lenovo, the world's largest personal computer maker, plans to equip its broad range of products - from PCs to smartphones and wearables - with AI technology. Earlier this month, it introduced Kira, a built-in cross-device intelligence system
that integrates with LLM partners.
"We are the only company besides Apple with significant market share across both PCs and mobiles, and in the open Android and Windows ecosystems," Lenovo CFO Winston Cheng told Reuters on the sidelines of the World Economic Forum in Davos.
Unlike Apple, which currently works only with OpenAI and Google's Gemini, Lenovo is looking to strike deals with many more LLM developers, he said.
Potential partners include Humain in Saudi Arabia, Mistral AI in Europe, and Alibaba and DeepSeek in China, according to Cheng.
"We're taking an orchestrator approach," he said. "We're not doing our own LLM. We're really doing a partnership because there are regulations around the world," said Cheng, a former tech investment banker who joined Lenovo in 2024 and became CFO in April 2025.
Asked about surging memory chip prices, which have weighed on the outlook for consumer electronics makers globally, Cheng said costs were rising and the company planned to pass these increases on to customers.
The technology company, which also makes servers, unveiled a partnership in January with U.S. AI chip leader Nvidia to help AI cloud providers quickly bring data centres into operation, via a liquid-cooled hybrid AI infrastructure.
Cheng told Reuters the two firms would focus on the "global deployment" of the capability and manufacture locally, and may consider launches in Asia or the Middle East.
(Reporting by Selena Li in Davos and Brenda Goh in Shanghai. Additional reporting by Casey Hall. Editing by Mark Potter)
