Capping a dramatic turnaround, Mythic has successfully raised $125 million in a fully oversubscribed funding round led by DCVC. The company is positioning itself to challenge the dominant GPU-based AI infrastructure with a revolutionary semiconductor architecture that delivers up to 100 times greater energy efficiency than today’s leading GPUs and competing AI ASICs. This leap in efficiency addresses the most pressing issue in modern AI: the massive, unsustainable energy consumption that plagues data centers and edge devices alike. Mythic’s technology stands out as the only semiconductor design to approach the energy efficiency of the human brain, enabling high-performance AI inference at a fraction of the power cost. This breakthrough makes it uniquely suited for applications where power and thermal constraints are critical—ranging from edge computing in smartphones and IoT devices to large-scale data center deployments. The company’s architecture leverages analog computing and a novel memory-centric design, eliminating the need for energy-intensive data movement between processors and memory. This fundamental shift allows Mythic to deliver real-time AI processing with minimal power, a key advantage as global demand for AI accelerates and energy costs rise. The oversubscribed round reflects strong investor confidence in Mythic’s potential to disrupt the AI hardware landscape. The funding will be used to scale production, expand its engineering team, and accelerate deployment of its chips in commercial and industrial applications. With AI workloads growing exponentially, the need for energy-efficient solutions has never been more urgent. Mythic’s technology not only promises to reduce the environmental footprint of AI but also opens new possibilities for AI in power-constrained environments—where traditional GPUs are simply too costly or impractical to deploy. As the AI industry grapples with the limits of current hardware, Mythic is emerging as a serious contender to reshape the foundation of artificial intelligence, one energy-efficient chip at a time.