1 min read
|
Saved February 14, 2026
|
Copied!
Do you care about this?
Nvidia has licensed AI-inference technology from the startup Groq, which specializes in chips designed for efficient AI processing. As part of the deal, Groq's CEO and some staff will join Nvidia, highlighting the increasing demand for advanced AI chips.
If you do, here's more
Nvidia has signed a licensing deal with the chip startup Groq to utilize its AI-inference technology. This agreement highlights the increasing demand for advanced AI chips in the market. As part of the deal, Groq's CEO Jonathan Ross, its president, and several staff members will join Nvidia. Founded in 2016, Groq focuses on creating chips and software designed for running AI models.
Groq's chips, known as language processing units, specialize in inference tasks, which involve AI models providing responses or predictions based on new data. Ross claims that Groq's chip design, which includes embedded memory, allows for quicker production and deployment while consuming less power than traditional graphics-processing units. These GPUs are known for their high energy consumption, especially during model training. This strategic partnership could enhance Nvidia's capabilities in the AI chip sector as competition intensifies.
Questions about this article
No questions yet.