About
Geekbench AI is a cross-platform AI benchmark that uses real-world machine learning tasks to measure performance on AI workloads. Geekbench AI measures the performance of your CPU, GPU, and NPU to determine if your device is ready for the cutting-edge machine learning applications of today and tomorrow.
Real AI Performance
Geekbench AI runs ten AI workloads, each with three different data types, providing a multi-dimensional picture of on-device AI performance. Using large datasets that simulate real-world AI use cases, both developers and consumers can measure on-device AI performance in just minutes using Single Precision, Half Precision and Quantized ratings.
Measure AI performance on CPU, GPU or NPU
Geekbench AI analyzes AI performance across the hardware stack - select your device's GPU, CPU, or dedicated NPU for testing. You can also choose available AI frameworks on your device, such as Core ML or QNN. Developers can determine the best combination of frameworks and models for specific workloads, and consumers can easily evaluate the impact of dedicated AI hardware.
Compare AI performance across platforms
Geekbench AI runs identical loads on Android, iOS, Windows, macOS and Linux. Our benchmark is designed for hardware with different capabilities - be it testing a smartphone with an ultra-energy-efficient NPU or a dedicated workstation with kilowatts of power for AI computing. Compare results instantly using our Geekbench AI Score Browser.