Announcing new capabilities that expand Google AI Edge Portal’s capabilities: benchmarking and debugging on-device LLMs. These new services give developers what they need to optimize gen AI performance accurately and efficiently across the entire Android ecosystem.

Google Cloud Platform provides a suite of cloud computing services for building, deploying, and managing applications and infrastructure on Google's global network. Developers can learn about cloud-native development, machine learning, and big data analytics to leverage GCP's scalable and reliable cloud infrastructure for their projects.

Google Cloud

Google AI Edge Portal now supports benchmarking and debugging on-device LLMs across a physical lab of over 120 Android devices. Developers can measure key metrics like initialization time, prefill speed, decode speed, and peak memory usage for LiteRT-LM format models on CPU and GPU backends. A newly integrated Model Explorer tool enables graph visualization, side-by-side model comparison, and per-layer analysis to help identify and fix conversion, quantization, and optimization issues. The feature is currently in private preview for allowlisted Google Cloud customers at no charge.

Benchmark LLMs on-device with AI Edge Portal

Benchmark LLMs across over 120 different mobile devices