![](https://techcrunch.com/wp-content/uploads/2022/09/GettyImages-1341418748.jpg?w=600)
Groq, a startup creating chips to run generative AI fashions quicker than typical {hardware}, has a watch towards the enterprise — and public sector.
At the moment, Groq introduced that it’s forming a brand new division — Groq Programs — centered on vastly increasing its buyer and developer ecosystem. Inside Groq Programs’ purview is serving organizations, together with authorities companies, that want to add Groq’s chips to current information facilities or construct new information facilities utilizing Groq processors.
In forming the brand new unit, Groq has acquired Definitive Intelligence, a Palo Alto–primarily based agency providing a spread of business-oriented AI options, together with chatbots, information analytics instruments and documentation builders. Definitive Intelligence CEO Sunny Madra is now main GroqCloud, Groq’s cloud platform that gives Groq {hardware} documentation, code samples and self-serve API entry to the corporate’s cloud-hosted accelerators.
“At Groq, we’re dedicated to creating an AI economic system that’s accessible and inexpensive for anybody with a superb concept,” Groq co-founder and CEO Jonathan Ross stated in a press launch. “We’re excited to welcome Sunny and his crew from Definitive Intelligence to assist us obtain this mission … The Definitive crew has experience in AI options and go-to-market methods, in addition to a confirmed dedication to sharing data with the group.”
Madra co-founded Definitive Intelligence in 2022 with Gavin Sherry, ex-director of engineering at EMC. Previous to launching Definitive Intelligence, Madra and Sherry co-launched Autonomic, a cloud-based platform for connecting mobility programs that Ford acquired in 2018.
Definitive Intelligence presents a number of business-oriented GenAI merchandise, together with OpenAssistants (a set of open supply libraries for creating AI chatbots) and Advisor (a visualization generator that connects to each enterprise and public databases). One among Definitive’s premier instruments is Pioneer, an “autonomous information science agent” designed to deal with numerous information analytics duties, together with predictive modeling.
Previous to the acquisition, Definitive Intelligence had raised $25.5 million in enterprise capital.
“The world is simply now realizing how vital high-speed inference is to generative AI,” Madra stated in an emailed assertion. “At Groq, we’re giving builders the velocity, low latency, and effectivity they should ship on the generative AI promise. I’ve been an enormous fan of Groq since I first met Jonathan in 2016 and I’m thrilled to affix him and the Groq crew of their quest to carry the quickest inference engine to the world.”
Groq, which emerged from stealth in 2016, is creating what it calls an LPU (language processing unit) inference engine. The corporate claims that its LPU can run current giant language fashions related in structure to OpenAI’s ChatGPT and GPT-4 at 10x the velocity.
Ross’ declare to fame helps to invent the tensor processing unit (TPU), Google’s customized AI accelerator chip used to coach and run fashions.
Definitive Intelligence is Groq’s second acquisition after Maxeler Applied sciences, a high-performance compute and AI infrastructure options agency, in 2022. It won’t be its final. The marketplace for customized AI chips is a extremely aggressive one, and — to the extent the Definitive buy telegraphs Groq’s plans — Groq is clearly intent on establishing a foothold earlier than its rivals have an opportunity.