Velaura AI Raises $110M for Power-Efficient AI Chip Design

Velaura AI has successfully closed a substantial $110 million Series A funding round, propelling its mission to engineer more power-conscious artificial intelligence processors. The significant capital infusion was spearheaded by Seligman Ventures, with crucial participation from Samsung Catalyst Fund and Mayfield, alongside a consortium of over half a dozen other investors. This funding achievement places the company's valuation north of the $1 billion mark, underscoring investor confidence in its innovative approach to AI compute economics.

The company, formerly known as Auradine and recognized for its high-performance cryptocurrency mining hardware, has strategically pivoted its focus. Velaura AI is now concentrating on its proprietary 'Titan Core' technology, a suite of processor building blocks and specialized services aimed at empowering clients to design their own AI chips. This strategic shift emphasizes a critical industry need: reducing the immense power consumption associated with advanced AI computations, particularly the matrix multiplications that underpin large language models.

According to Velaura AI, these core AI operations can consume up to 70% of a chip's total energy budget. Their Titan Core solution claims to slash the energy required for these calculations by a factor of two to four. This efficiency gain translates into tangible cost savings, with the company estimating potential reductions of up to $1,300 per chip over a three-year period. This focus on 'compute economics' is becoming paramount as AI models grow in complexity and demand for ubiquitous AI deployment increases.

Rajiv Khemani, CEO of Velaura AI, articulated the company's vision, stating, "Every advance in AI, from reasoning models to embodied intelligence, creates demand for more compute, and ultimately more power. The next era of AI will be defined not only by better models, but also by fundamentally better compute economics." The company's offering includes an optimized cell library, enabling chip designers to bypass the foundational work of creating basic transistor sets. For more sophisticated needs, Velaura AI also facilitates the creation of custom low-voltage circuits, requiring only an RTL file blueprint from clients.

Velaura AI's technology supports cutting-edge manufacturing processes, including the latest three- and two-nanometer nodes. The company revealed it is actively collaborating with several hyperscalers on chip projects utilizing these advanced fabrication technologies. Beyond large-scale data center applications, Velaura AI anticipates its low-voltage chip designs will find significant adoption in edge devices, further broadening the reach of efficient AI processing.

The newly acquired capital will be strategically deployed to accelerate Velaura AI's engineering roadmap and expand its customer-facing teams. This investment arrives at a pivotal moment for the semiconductor industry, as the insatiable demand for AI processing power clashes with growing concerns over energy consumption and sustainability. Velaura AI's emphasis on power efficiency positions it as a key player in addressing these dual challenges, potentially reshaping the economics of AI hardware development and deployment.