AI BriefingGartner Highlights FuriosaAI in Power-Efficient AI Chip Sector
Key point
Gartner stated that FuriosaAI's power-efficient chips deserve attention for reducing operational costs when adopting AI hardware.
Details
According to Gartner's latest report, the key metric for generative AI hardware adoption is not simple performance (TOPS, FLOPS) but ROI (return on investment). Companies are considering energy efficiency, which determines operational costs, as the most important factor when adopting AI.
In this report, Gartner mentioned FuriosaAI as a notable sample vendor alongside established giants like NVIDIA, Intel, and AMD. In particular, FuriosaAI's second-generation chip, RNGD, is optimized for LLM and multimodal model inference, and was presented as an alternative that addresses the surging power consumption problem by offering superior power efficiency compared to existing GPUs.
Currently, high-performance GPUs consume over 700W of power and are expected to reach 2,000W in the future. Accordingly, companies must simultaneously address CAPEX (capital expenditure) and OPEX (operational expenditure) issues by choosing products with the optimal combination of performance, power requirements, and cost.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.