This repository hosts a large, verified collection of pre-trained state-of-the-art models for use with the ailia SDK, a cross-platform inference runtime supporting GPU acceleration via Vulkan and Metal on Windows, Mac, Linux, iOS, Android, Jetson, and Raspberry Pi. Models span computer vision tasks (object detection, segmentation, pose estimation, face recognition), language tasks (LLMs, embeddings, tokenizers), and audio tasks (speech recognition, voice synthesis). The repository includes sample code and tutorials for running each model.