NVIDIA Vera CPU Claims 2x Agent Speed and 1.6x More Agents Per Watt
NVIDIA says Vera beats leading x86 CPUs on AI agent speed and efficiency, with partner tests showing up to 2.2x gains and rapid sandbox scaling at scale.
Summary
On November 5, 2026, NVIDIA positioned Vera as a single CPU for power-limited AI factories, where agents repeatedly move between GPU reasoning and CPU-based searches, database queries, coding, testing and actions. In NVIDIA’s 100-megawatt data center comparison, Vera delivered industry-leading agent speed and 1.6x more agents per watt than the fastest x86 CPU. Against the most energy-efficient x86 CPU, it ran agents 2x faster while retaining leading efficiency, potentially replacing separate fast and dense x86 parts.
Early production-style tests supported those claims. Perplexity ran agentic coding tasks 1.5x faster than on traditional CPUs and plans a production deployment. DeepInfra reached up to 2.2x the speed of the best x86 processor for production agent orchestration, leading four tested architectures. ClickHouse’s standard binary, without Vera-specific tuning, scaled near-linearly across 176 cores and topped the ClickBench hardware leaderboard. DuckLabs achieved about 1.5x the median TPC-H SF1000 performance of an Intel Xeon 6 baseline. Daytona started 1,000 agent sandboxes in 10.8 seconds versus 88.4 seconds on x86 and completed 12.7 jobs per second with 2,000 concurrent sandboxes. Modal delivered up to 2x more sandbox throughput per core than comparable x86 hardware, with larger gains under heavy load.
Positives
- Vera supported 1.6x more agents per watt than the fastest x86 CPU in NVIDIA’s 100-megawatt comparison.
- Vera ran agents 2x faster than the most energy-efficient x86 CPU while maintaining leading energy efficiency.
- DeepInfra recorded up to 2.2x the best x86 speed and led all four architectures tested.
- Daytona started 1,000 sandboxes in 10.8 seconds, compared with 88.4 seconds using x86.
- ClickHouse scaled near-linearly across 176 cores and topped ClickBench without Vera-specific tuning.
- Modal achieved up to 2x sandbox throughput per core, with gains increasing under heavy load.
Risks & concerns
- NVIDIA’s 100-megawatt comparison does not identify the fastest or most energy-efficient x86 CPU models used.
- Vera pricing, general availability and deployment timelines were not provided.
- Perplexity’s 1.5x result precedes its planned production deployment.
- The partner results are described as production-style tests, leaving broader at-scale production performance unspecified.
- Detailed benchmark configurations and absolute power consumption figures were not disclosed.
