The core of an effective AI system lies in having a thoroughly trained model. Achieving this necessitates substantial resources and energy, making it potentially expensive for stakeholders including investors. However, one emerging company is disrupting the norm by utilizing cost-effective equipment for their training process. It’s said that Fastino is employing under $100,000 worth of gaming GPUs to meet their training requirements. In contrast with conventional business models such as
Elon Musk’s giant xAI,
This requires extremely high energy consumption.
TechCrunch
first reported
on Fastino and its use of low-end gaming chips.
As per Fastino, they achieved this by crafting an architecture tailored specifically for their AI model, which is built to operate efficiently at a reduced scale. This adjustment resulted in a speedier model with significant precision. Nonetheless, the most impressive aspect is the economical advantage gained from using affordable gaming equipment for training their system.
Despite having an investment of under $100K in GPUs, Fastino has begun generating significant funds from investors who recognize its potential. Notably, the company secured $17.5 million from Khosla Ventures most recently, building upon the $7 million it garnered towards the end of last year through investments from M12, Microsoft’s venture arm, along with Insight Partners.
Tiny Corp is not the sole enterprise pushing boundaries in AI hardware with budget-friendly development solutions. They have introduced their take on this with something they refer to as
TinyBox
Last year, using six repurposed AMD Radeon 7900 XTX graphics cards specially designated for AI tasks.
Up until now, Fastino hasn’t provided any performance metrics, making it impossible for us to verify precisely how effective their latest architectural design is compared to rival AI systems. They have issued comments indicating significant capabilities, emphasizing its capacity to deliver intricate responses almost instantaneously. However, lacking additional details makes it challenging to assess its competitiveness with leading players in the artificial intelligence sector.
Follow
Tom’s Hardware on Google News Feed
to get our up-to-date news, analysis, and reviews in your feeds. Make sure to click the Follow button.
Enjoying this article? To discover more articles like this one, follow us on MSN by clicking the +Follow button at the top of this page.
