金色财经|Jul 21, 2026 16:21
[NVIDIA Details Next-Generation AI-Specific Vera CPU, Claims Speed is Over Twice That of Other CPUs]
According to a report by Jinse Finance, on July 22, NVIDIA published an article stating that benchmark results from DeepInfra show the NVIDIA Vera CPU is more than twice as fast as other CPUs and can support more concurrent AI agents. Cloud platform DeepInfra, an early participant in NVIDIA's open AI ecosystem, independently designed and conducted benchmarks based on its production AI agent infrastructure. DeepInfra processes nearly five trillion tokens weekly, approximately 30% of which are driven by agent systems. Its cloud platform is built specifically for high-throughput AI inference.
The benchmark results indicate that, under the same quality of service, it supports up to 1.6 times more concurrent AI agents, achieves collaborative speeds up to 2.2 times faster, and improves infrastructure utilization and cost efficiency. These findings demonstrate that the NVIDIA Vera CPU can deliver the cost-effectiveness, low latency, and throughput required for production-grade AI agents.
As AI agents take on more complex reasoning, planning, tool usage, and data flow tasks, CPU performance is becoming increasingly critical for the coordination work surrounding each model invocation. As part of NVIDIA's extreme co-design strategy for AI factories, the Vera CPU is specifically designed for agent-based workloads. DeepInfra's benchmarks highlight how the NVIDIA Vera CPU helps cloud service providers enhance infrastructure utilization, improve cost efficiency, and support more concurrent AI agents under the same quality of service.
Share To
Timeline
HotFlash
APP
X
Telegram
CopyLink