Startup goes after a fragmented inference AI market where there are no general-purpose inference accelerators capable of processing AI workloads across different applications.
What’s at stake?
Armed with its own unique at-memory compute architecture, Untether AI wants to lead the general-purpose AI inference accelerator market. Can the startup unseat leading CPU and GPU vendors that dominate the AI training sector and are extending their reach into AI inference? Is the knock-your-socks-off performance enough to pull that off?