Nvidia Corp. CEO Jensen Huang said Wednesday that the company's A100 fleet is "mission-capable from 2020 through 2029," highlighting the longevity of its data center GPUs. The A100 debuted in 2020 as part of Nvidia's Ampere generation. "NVIDIA computing is more than chips," Huang said on X, pointing to the CUDA software platform as a key reason its hardware can remain useful across generations. "CUDA makes NVIDIA computing versatile. Versatility makes it fungible. Fungibility drives utilization and extends durability." His comments followed a disclosure from AI cloud provider CoreWeave Inc. that it signed a contract to rent Nvidia A100 GPUs through 2029 at what CFO Nitin Agrawal described as an "attractive price." The development challenges concerns that accelerating chip improvements could cause older GPUs to lose economic value within two or three years. Older GPUs can remain useful by shifting from demanding AI workloads to inference, smaller models and other applications, according to a Business Insider report. Rental-market data also supports CoreWeave's assessment, with Silicon Data saying Tuesday that A100 rates have remained resilient following a strong rebound in 2026. Silicon Data added that based on residual fair value estimates, A100 stopped depreciating since late 2025 as rising rental demand supports its value.