As part of a bigger tech alliance, Nvidia has bought $5 billion worth of Intel shares.
Intel revealed the purchase of 214.7 million shares in a filing with the Securities and Exchange Commission on Monday, December 29. This was the end of a deal that was first announced in September.
Reuters reports that the purchase is a big help for Intel after years of mistakes and expensive expansions of its production capacity that drained its resources.
The statement comes just a few days after the Federal Trade Commission (FTC) said that authorities had given the green light to the proposed investment. This is seen as a big vote of confidence in Intel by the world’s most valuable corporation.
The two businesses are also starting a cooperation focused on product development. They want to work together on several generations of specialized data center and personal computing solutions to “speed up applications and workloads across hyperscale, enterprise, and consumer markets.”
Nvidia also recently bought Groq, a company that makes custom-built inference processors, and hired several of its employees. In a statement to PYMNTS last week, a corporate spokeswoman said, “We haven’t bought Groq.” We hold a non-exclusive license to Groq’s IP, and we have engaged engineers from Groq’s team to help us achieve our goal of providing the best accelerate computing technology in the world.
Groq said in a blog post that the firms want to make it easier for people to get high-performance, low-cost inference, which is the stage in which a trained model evaluates fresh data and gives findings. Inference is at work whether a customer care chatbot answers a question or an AI system looks at a financial document.
“Training makes the model by processing huge amounts of data to find patterns. Inference uses that knowledge to do specific tasks on a large scale,” PYMNTS said earlier this year. “As businesses use AI systems to handle thousands or millions of requests every day, inference becomes the biggest operational problem and cost driver.”
This month, Nvidia also released its Nemotron 3 family of open models. These models are meant to provide transparent, efficient, and specialized agent-centric AI across several industries.
As said here, developers and businesses may use these models, data, and tools to make and change AI agents that can do things like code and reason, as well as automate complex workflows.
The paper added, “Open-source AI models are large pretrained neural networks whose weights and code are publicly available for download, inspection, modification, and redistribution.”
“Unlike closed or proprietary models that are only available to one provider, open models let developers, researchers, and businesses change the model to fit their needs, check how it works, and add the technology to their systems without having to worry about licensing issues.”
Read more AI News here.
