AI Benchmark Mobile

Version:  5.1.1   |   Last Updated: 06.01.2024

AI Benchmark V5 is designed for the next-generation mobile AI accelerators, and is introducing numerous new tests and workloads including FullHD and 4K video super-resolution, real-time question answering, RAW image processing, the latest neural networks for image recognition, photo reconstruction and NLP tasks, and even power consumption measurements. Get it now to extensively evaluate your device!

AI Benchmark Nightly

Last Updated:  06.01.2024

AI Benchmark Mobile version built with the latest TensorFlow Lite nightly runtime, with support for the full range of TensorFlow ops and with the newest Android TFLite delegates: GPU, NNAPI, Hexagon and Neuron. Allows developers to test their own TFLite models converted with TF-nightly or containing custom TF operators. Provided for R&D only, might not be compatible with some devices. XNNPACK has multithreading issues.

AI Benchmark Mobile

Version:  4.0.4

Is your smartphone capable of running the latest Deep Neural Networks to perform these AI-based tasks? Does it have a dedicated AI Chip? Is it fast enough? Run AI Benchmark to professionally evaluate its AI Performance!

AI Benchmark Mobile Lite

Version:  4.0.4 Lite

Designed for legacy devices running Android 4.x or having less than 1Gb of RAM. Contains a reduced number of tests, supports CPU-based inference only. The scores are calibrated to be comparable with the results of the full version.

AI Benchmark Alpha

Version:  0.1.2

AI Benchmark Alpha is an open source python library for evaluating AI performance of various hardware platforms, including CPUs, GPUs and TPUs. The benchmark is relying on TensorFlow machine learning library, and is providing a precise and lightweight solution for assessing inference and training speed for key ML models.

Burnout Benchmark

Version:  2.0.2

Burnout Benchmark is a professional tool for an extensive performance evaluation of all chipset components. It pushes the device to its limits by loading CPU, GPU, NPU and DSP by 100% separately or simultaneously to assess the real-world device performance under heavy load and check its thermal throttling behavior over time.