ArchestackBest of Open-Source AI: List on GitHub ↗ (opens in a new tab)

#161161st of 206 in ML frameworks and libraries

lite.ai.toolkit

C++ library running 100+ pretrained vision models via ONNXRuntime, MNN and TensorRT

Repository on GitHub ↗ (opens in a new tab)
Stars
4.4k
License
GPL-3.0
Last commit
Sep 2026
Last release
Jul 2025
Language
C++

Missed the latest GitHub check; these facts may be older.

Overview

Lite.Ai.ToolKit is a C++ library that wraps pretrained computer-vision models (YOLO variants, face detection and recognition, segmentation, matting) behind a consistent API such as lite::cv::detection::YoloV5. It runs on ONNXRuntime by default, with optional MNN and TensorRT (10.x, CUDA 12.x) backends. The README claims 300+ C++ implementations and 500+ weights, distributed through ONNX, MNN, NCNN and TNN model hubs.

Who it is for: C++ developers embedding vision model inference in native apps

Strengths

  • Default dependencies are only OpenCV and ONNXRuntime
  • Same C++ call pattern across detection, face, segmentation and matting models
  • Optional TensorRT backend for YOLO, face models, StableDiffusion and FaceFusion
  • Prebuilt Linux library and sample weights published on the v0.2.0 release

Weaknesses

  • Build script officially supports Linux only; macOS and Windows are marked unofficial
  • GPL-3.0 license restricts use in closed-source products
  • Last release was 2025-07; maintainer says his focus moved to LLM inference
  • Mostly vision models; no LLM serving or text generation

What it needs

  • GPU optional
  • Needs OpenCV, ONNXRuntime 1.17.1, MNN 2.8.2 (optional), TensorRT 10.x (optional), CUDA 12.x (optional)
  • Models: ONNX, MNN, NCNN, TNN, TensorRT engine

Also in ML frameworks and libraries

See all 206
Also in ML frameworks and libraries
RankProjectScore
1transformersPython library defining open models for text, vision, audio and multimodal tasks168.9k stars+1,573 (+0.9%) in 30 daysApache-2.094 out of 100
2ultralyticsPython library and CLI for training and running YOLO vision models63.3k stars+1,004 (+1.6%) in 30 daysAGPL-3.094 out of 100
3pytorchPython tensor library with GPU support and tape-based autograd105.5k stars+1,036 (+1.0%) in 30 dayscustom license93 out of 100
#4pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts37.3k stars+107 (+0.3%) in 30 daysApache-2.093 out of 100
#5supervisionPython toolkit for loading, annotating and counting computer vision detections Live demo ↗ (opens in a new tab)51.6k stars+521 (+1.0%) in 30 daysMIT92 out of 100