9,600 Nvidia A100 GPUs Listed on eBay for $59.7M
A Minnesota-based used IT hardware seller listed a decommissioned AI cluster containing 9,600 Nvidia A100 GPUs on eBay for $59.7 million, with shipping included. The listing ended on October 28, and the seller did not say whether the equipment had sold.
The A100 GPU cluster comprised 300 racks and 1,200 servers, with AMD EPYC processors and about 269,000 CPU cores. Its GPU count exceeds the 6,080 A100s used in the first phase of Meta’s AI research supercomputer. The seller’s asking price works out to roughly $6,200 per A100 when the full system is included. Tom’s Hardware estimated that the GPUs, processors and memory alone were worth about $28.1 million, excluding infrastructure such as racks, cooling and power equipment.
Despite being based on a 2020-generation GPU, the A100 cluster may still appeal to AI computing providers. CoreWeave’s chief executive said the company had signed an A100 rental contract through 2029, while older GPUs can offer a lower-cost alternative to newer systems facing supply delays. The listing’s removal leaves its sale status unknown.
Neutral
The listing has no clear, direct effect on cryptocurrency prices or market stability. It concerns the resale of a large AI computing cluster, rather than crypto assets, token markets or blockchain networks. Although GPU supply and pricing can matter to crypto miners, the Nvidia A100 is a data-centre accelerator used mainly for AI workloads, and the article provides no indication that this equipment will be redirected to mining.
In the short term, traders may view the sale as another sign of ongoing demand for AI computing capacity and continued value in older hardware. That could support sentiment around AI-related technology companies, but it is not a reliable signal for crypto prices. The listing’s removal is also inconclusive: the equipment may have sold, become unavailable, or been withdrawn for another reason.
Over the longer term, continued use of older GPUs could affect the cost and availability of AI compute, while investment in newer systems may influence data-centre spending and semiconductor demand. These trends can shape broader risk appetite and technology-sector sentiment, which sometimes spill over into crypto markets. However, as in past instances where major AI hardware announcements produced sector-specific reactions without a consistent move across digital assets, any crypto impact here is likely indirect. Traders should look to wider indicators—such as Bitcoin price action, liquidity, macroeconomic conditions and crypto-specific flows—rather than treat this listing as a market catalyst.