# xlite-dev/lite.ai.toolkit

A lite C++ AI toolkit: 100+ models with MNN, ORT and TRT, including Det, Seg, Stable-Diffusion, Face-Fusion.

Repository: https://github.com/xlite-dev/lite.ai.toolkit
Canonical: https://ross.abutalabs.com/products/liteaitoolkit
Homepage: https://github.com/xlite-dev/lite.ai.toolkit
Language: C++
License: GPL-3.0
License Family: copyleft
Topics: yolox, onnxruntime, robustvideomatting, mnn, ncnn, tnn, yolov6, onnx, yolov5, yolov8, tensorrt, mnn-model, facefusion, stable-diffusion
Last push: 2026-08-26T09:14:20+00:00

## Health v2 (maintenance only)
Score: 74/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 28, longevity 100
- inputs: {"age_days": 1990, "days_push": 7, "days_rel": 428, "gap_med": 56.0, "n_releases_24m": 7}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4427, forks 785 (observed 2026-08-28T04:08:48.920008+00:00)

## What it is
A lightweight C++ toolkit providing unified APIs for 100+ pre-trained AI models across inference backends like ONNX Runtime, MNN, TensorRT, TNN, and NCNN. It covers object detection, segmentation, face detection/recognition, matting, Stable Diffusion, and Face Fusion.

## Use cases
- run yolov5/yolov8 object detection in C++
- deploy face detection and recognition models natively
- run stable diffusion image generation in C++
- perform image matting and background removal
- swap inference backends between onnxruntime, mnn, and tensorrt
- port python AI models to C++ production applications

## When to choose
- you need C++ inference for many vision models with a consistent API
- you want to deploy models on CPU or GPU without Python dependencies
- you need multiple backend support (ORT, MNN, TRT, NCNN, TNN) in one toolkit

## When to avoid
- you need training or fine-tuning capabilities
- you work primarily in Python and don't need C++ integration
- you need a model not covered by its model zoo

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, computer-vision, llm-inference, stable-diffusion
- domain: computer-vision, machine-learning, deep-learning, image-processing, artificial-intelligence
- platform: cpp, cross-platform
- tags: onnxruntime, tensorrt, mnn, ncnn, tnn, yolov8, object-detection, face-recognition, segmentation, matting, face-fusion, model-zoo, inference, linux, gpu

## Member repositories
- xlite-dev/lite.ai.toolkit (main) score 74

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:48.920008+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:20:56.809883+00:00, confidence not recorded.
  - readme: https://github.com/xlite-dev/lite.ai.toolkit (fetched 2026-08-28T04:08:48.920008+00:00, sha bc3156bd81e6)
  - homepage: https://github.com/xlite-dev/lite.ai.toolkit (fetched 2026-08-29T09:08:20.357926+00:00, sha 6617643e6147)
- Data as of 2026-08-30T08:39:29.467469+00:00.