On Device Verification

Prove a converted or quantized LiteRT model on the actual device via the CompiledModel API - confirm GPU residency, compare device output against the source model, and diagnose device-only failures such as silent CPU fallback, whole-graph compile ceilings, and fp16 range breaks. Use after conversion or quantization, when device output is wrong or NaN, when a clean graph fails to compile only on device, or when GPU and CPU outputs are suspiciously identical.

google-ai-edge Updated 25.3k repo stars

File contents

google-ai-edge/litert-samples commit 5f725441f4

Frequently asked questions

npx get-skill add google-ai-edge/on-device-verification