Cross-platform inference engine supporting GPU inference via Metal and Vulkan. A model library with over 400 types, advanced pre/post-processing, and bindings for Python, C++, C#, and Flutter.
Comprehensive API references and guides for each component of the ailia SDK ecosystem.
ONNX inference API for high-performance AI model execution on any platform.
TensorFlow Lite compatible runtime for lightweight model inference.
Local large language model inference with GGUF format support.
Speech recognition engine for accurate Speech-to-Text conversion.
Text-to-speech synthesis engine for natural voice generation.
Object movement tracking for real-time video analysis and surveillance.
Tokenization library for NLP text encoding and decoding tasks.
AI inference library that runs machine learning models in the browser using WebAssembly.
Resources, downloads, and additional information about the ailia SDK.
