Skip to content

usls

A cross-platform Rust library powered by ONNX Runtime for efficient inference of SOTA vision and vision-language models.


✨ Features

  • High Performance


    Multi-threading, SIMD, and CUDA-accelerated processing with ONNX Runtime execution providers

  • Cross-Platform


    Linux, macOS, Windows with support for CUDA, TensorRT, CoreML, OpenVINO, DirectML

  • Unified API


    Single Model trait with run()/forward() and unified Y output for all models

  • Auto-Management


    Automatic model download from HuggingFace/GitHub, caching and path resolution

  • Multiple Inputs


    Support for images, directories, videos, webcam, streams and combinations

  • Precision Support


    FP32, FP16, INT8, UINT8, Q4, Q4F16, BNB4, and more quantization options

  • Full-Stack Suite


    Complete workflows with DataLoader, Annotator, and Viewer

  • Model Ecosystem


    50+ SOTA vision and VLM models ready to use