Nvidia has signed a $6 billion deal with AI startup Poolside to develop a powerful open-weight model capable of competing with Chinese LLM such as DeepSeek and Kimi K3. This was reported by WSJ citing its own sources.
On August 21, Bloomberg reported that the chipmaker agreed to pay Poolside $6 billion for AI model licenses, offer jobs to more than 100 startup employees, and invest an additional $1 billion. According to the article, Nvidia has recently focused on supporting open-source models that can be freely used and customized by developers. The company aims to make the technology more accessible and prevent concentration in the hands of a few major players. Nvidia also links these efforts to supporting future demand for its chips and AI systems.
The deal with Poolside is structured similarly to Nvidia’s recent acquisition of chip developer Groq. Reports indicate the company paid $20 billion for the startup’s licenses and hired most of its employees and founders.
According to WSJ, Nvidia’s agreement with Poolside will also pose a direct challenge to leading American AI companies, including OpenAI and Anthropic, as open-weight models are generally much cheaper to operate and allow for easy customization.
In July, it was reported that the administration of U.S. President Donald Trump resumed discussions on measures that could complicate the use of Chinese open-weight models. Twenty-five American companies, including Nvidia, Meta, and Microsoft, urged the authorities not to impose restrictions. In an open letter, they warned that bans would not strengthen the United States’ technological leadership but rather shift the market towards a few closed developers.
The release of Kimi K3, whose creators the White House accused of distilling Fable from Anthropic, largely intensified the discussion regarding potential restrictions.
Earlier, ForkLog explored the phenomenon of the Chinese model from Moonshot AI in an article.
