‘Use AI models for as little as $5,000’: Brave CPU startup that claimed it saved 99% on AI costs now wants to sell you an AI workstation – with an incredible price tag and 1TB RAM
CPU startup Tachyum has previously said that one of its Prodigy Universal Processor units could rival dozens of Nvidia H200 GPUs. The 192-core 5nm processor delivers 4.5x the performance of the best processors for cloud workloads and claims to be six times more effective than GPUs for AI.
Now the company has announced the Prodigy ATX Platform, an AI workstation that promises to run advanced AI models for just $5,000. The ambitious system, which features an impressive 1TB of memory, is designed to make advanced AI models accessible to a wider audience.
The Prodigy ATX platform is built around a 96-core Prodigy processor, which is designed with only half of the chip enabled, a strategy aimed at reducing power consumption and increasing yield, reducing costs and the platform becomes more accessible.
Effective for LLMs
The system is expected to be equipped with 1TB DDR5-6400 SDRAM with 16 memory modules, offering a peak bandwidth of 819.2 GB/s. The system design includes three PCIe x16 5.0 slots, three M.2-2280 NVMe slots with a PCIe 5.0 x4 interface and SATA connectors for SSDs and HDDs.
Tachyum says its Prodigy ATX platform is particularly effective for running LLMs that are notorious for their high memory capacity requirements.
A single Prodigy system can reportedly run a ChatGPT4 model with 1.7 trillion parameters, requiring 52 NVIDIA H100 GPUs to run at significantly higher costs and power consumption.
Despite the impressive specifications, there are doubts about the economic viability of the Prodigy ATX Platform for Tachyum. If Tom’s hardware points out that the total cost of the system components, excluding the Prodigy processor, is estimated at approximately $4800. Tachyum CEO Dr. Radoslav Danilak, however, remains optimistic, stating that the platform’s powerful AI capabilities will enable organizations of all sizes to compete in AI initiatives.
“Generative AI will reach widespread use much sooner than anyone initially expected,” said Dr. Danilak. “In a year or two, AI will be a required component on websites, chatbots and other critical productivity components to ensure good user experience. Prodigy’s powerful AI capabilities will allow LLMs to operate much more easily and cost-effectively than existing CPU + GPGPU based systems, allowing organizations of all sizes to compete in AI initiatives that would otherwise be dominated by the largest players in their industries.”
The launch of the Prodigy ATX Platform has been postponed multiple times, with the latest plan pegging the processor’s launch for the second half of 2024. It remains to be seen if Tachyum can deliver on its promises and revolutionize AI landscape.