The partnership between Red Hat Inc. and Cisco Systems Inc. has strengthened in the era of artificial intelligence, as shown by Red Hat’s announcement of llm-d.

The llm-d community was made to handle generative AI inference at scale. Cisco partnered with Red Hat on the project, extending its support for AI development using Red Hat’s open-source tools.

The set of theCUBE at Red Hat Summit 2025.

“You basically harness that immense power of an open-source ecosystem and all those contributors, but you harden it, you support it for the enterprise, and so [Red Hat] plays a critical role, and they’ve played a critical role for a long time,” said Todd Brannon (pictured, left), senior director of ecosystems solutions at Cisco. “They’ve been woven into all of our solutions over time. Originally, with networking, compute as we got into that, security. They’re really intrinsic to all the solutions that we’ve had.”

Brannon and Chris O’Brien (pictured, right), senior technical marketing director at Cisco, spoke with theCUBE’s host Rebecca Knight and theCUBE Research’s Rob Strechay at Red Hat Summit, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed the llm-d project, as well as other AI-focused collaborations. (* Disclosure below.)

Doing AI at scale through llm-d and more

As Cisco looks to build out infrastructure for large-scale enterprise IT, Red Hat’s suite of open source solutions has been a boon. The llm-d project, in particular, allows for more efficient and scalable inference by integrating advanced virtual large language model-based inference capabilities into existing enterprise IT infrastructures, according to Brannon.

“Inference is the point of delivery for all this AI work,” he said. “Modern applications are these constellations of microservices, and so being able to scale out horizontally is essential. That’s what that llm-d piece is doing for us. [It] help(s) us spread this across multiple nodes … It’s Red Hat taking the lead to bring really important capabilities in to help our enterprise customers scale.”

The rise of AI models has led some companies to seek more on-premises solutions or build their own private AI clouds for improved security. In response to this trend, Cisco has acquired Isovalent, Inc., a cloud native networking and security company. The goal, O’Brien emphasizes, is to give customers what they want.

“For us at Cisco, just to be point-blank, we want to enable our customers to go where they want to go,” he said. “And if it happens to be OpenShift virtualization or they want to stay with Broadcom or what, we want to bring the same consistent approach to the fabric as well as to the compute.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of Red Hat Summit:

(* Disclosure: Cisco Systems Inc. sponsored this segment of theCUBE. Neither Cisco nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE