Pre-converted Models models-v1.0.0

Pre-converted AI Models for ai_benchmark

These models are pre-converted for optimal performance on each platform:

TFLite Models (Mobile/macOS/Windows/Linux)

• Mobile GPU acceleration via Metal (iOS/macOS) or OpenCL (Android)
• Desktop: TFLite CPU fallback available

ONNX Models (Windows/Linux, if converted)

• GPU acceleration via DirectML (Windows) or CUDA (Linux)

Usage

The ai_benchmark app will automatically download the optimal format for your platform.
If ONNX is unavailable, TFLite will be used as fallback on Windows/Linux.

Pre-converted Models models-v1

Pre-converted AI Models for ai_benchmark

These models are pre-converted for optimal performance on each platform:

TFLite Models (Mobile/macOS/Windows/Linux)

• Mobile GPU acceleration via Metal (iOS/macOS) or OpenCL (Android)
• Desktop: TFLite CPU fallback available

ONNX Models (Windows/Linux, if converted)

• GPU acceleration via DirectML (Windows) or CUDA (Linux)

Usage

The ai_benchmark app will automatically download the optimal format for your platform.
If ONNX is unavailable, TFLite will be used as fallback on Windows/Linux.