CoreWeave extends Nvidia A100 contracts to 2029, proving older AI GPUs still generate profit

CoreWeave proves Nvidia's aging AI GPUs from 2020 can generate profit nine years after deployment, signs A100 contracts into 2029 -- power constraints and legacy infrastructure keep old GPUs profitable

CoreWeave signed a contract for Nvidia A100 GPUs that runs into 2029, nine years after the Ampere-based SKU debuted, CEO Mike Intrator disclosed on the company's second-quarter earnings call on Monday. The company reported $2.58 billion in quarterly revenue, up 112% year over year, and a $104 billion revenue backlog that excludes more than $25 billion in new commitments booked since July. In remarks made during the earnings call, Intrator said, "pricing for prior generation SKUs is at or above where it was years ago." Michael Burry accused hyperscalers in November of understating depreciation by $176 billion between 2026 and 2028 by stretching GPU useful-life assumptions to five or six years against a chip cadence that now delivers a new Nvidia architecture roughly every year. Concerns that accelerating GPU obsolescence could unravel AI financing predate his estimate, and an unnamed Google architect put datacenter GPU service life at one to three years back in 2024. Intrator told CNBC at the time that a batch of H100s coming off an expired contract was immediately rebooked at 95% of the original price, and Nvidia CFO Colette Kress countered Burry directly, saying A100s sold six years ago still run at full utilization. The 2029 deal stretches that record further, putting contracted revenue on 2020 silicon beyond even the six-year depreciation schedules defended by hyperscalers. An air-cooled Nvidia DGX A100 system draws 6.5kW at maximum load and fits comfortably in legacy data center halls designed for roughly 20kW per rack. Nvidia's current GB200 and GB300 NVL72 racks draw 120kW to 140kW and require direct-to-chip liquid cooling, roughly six times what those older facilities can feed and cool. Blackwell therefore can't move into the halls where Ampere is currently deployed without a rebuild of power delivery and cooling infrastructure. That mismatch is helping to keep A100 fleets in service regardless of how gracefully or not silicon itself ages. The energized, air-cooled capacity holding them has no higher-value use, and renting nine-year-old GPUs is a better alternative to leaving it in the dark. CoreWeave's contracted power grew to 3.7 GW in Q2 and stood at 4.2 GW as of Monday, against just 1.5 GW online, so customer commitments already cover nearly triple the capacity the company can deliver. CFO Nitin Agrawal added on the call that capacity coming up for renewal represents a very limited share of CoreWeave's fleet, with older-generation ASPs at or above levels from a year ago. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.

NVIDIA's Six-Year-Old A100 Defies Deprecation As CoreWeave Rents It Through 2029, Stretching Ampere To A Nine-Year Lifespan

CoreWeave has committed to a rental plan for NVIDIA's A100 GPUs through 2029, highlighting how CUDA upgrades make older platforms viable for extended durability. NVIDIA A100 Might Be Two Generations Old, But CoreWeave Has Found A Use For It In AI Through 2029 The NVIDIA A100 AI accelerators are part of the Ampere GPU generation, which launched back in 2020. The 2nd Generation Tensor Core chip has been out for six years, and by this time, we would usually expect it to be phased out, but that's not the case as Jensen implies that NVIDIA's chips are designed for extended durability as CUDA upgrades make even older compute platforms a viable and a "productive asset" that is rentable, durable, and financeable for long-term use cases. This statement comes after CoreWeave announced that it will rent out NVIDIA's A100 GPUs through 2029. Using older GPUs comes at a time when shortages and price hikes have affected almost every segment of the tech world. We have even seen much older GPUs such as the Volta-based V100 series seeing a bump in prices. In a previous report, we covered how rental prices for A100 and H100 GPUs were at an all-time high. Besides the shortages, it should also be noted that continued efforts to optimize new AI models and upgrades within the CUDA ecosystem have made it possible to run workloads on the same amount of compute rather than upgrading to newer hardware, which requires much higher investments. There has been a debate about hardware deprecation considering just how accelerated AI roadmaps have become. Both NVIDIA and AMD are now following a yearly cadence, which means that a better and more productive chip will be out each year for AI workflows. And it's not always the best approach to invest billions into the current generation when a new generation is close by. With these optimizations and extended lifecycles, existing GPUs still have a lot of room to breathe. A 2029 commitment to A100s means that it will serve for 9 years. CoreWeave also mentions that they were able to get an attractive price for NVIDIA's A100 chips. The Blackwell and Rubin chips offer much more advanced compute capabilities but come at a massive price. So this goes on to show that AI accelerators have a much longer shelf life than most estimated. And it's a good sign, as large inventories of unused GPUs can be put to work instead of gathering dust, and reduce stress on the overall supply chain. News Source: Business Insider Follow Wccftech on Google to get more of our news coverage in your feeds.