open-edge-platform/geti

Build, train, optimize, and run computer vision models locally, from raw images to live inference. Open source, optimized for Intel XPU (CPU-only and CUDA also supported).

What it solves

Geti is an end-to-end framework designed to simplify the entire lifecycle of computer vision AI models. It removes the friction between raw image collection, data annotation, model training, optimization, and final deployment, particularly for edge devices.

How it works

Geti provides two primary interfaces: a no-code web application for rapid prototyping and the getitune Python library for programmatic control. The system allows users to annotate images using AI-assisted tools (like the Segment Anything Model), train models from a vast catalog of state-of-the-art architectures (including YOLO and DETR variants), and optimize them using OpenVINO for high-performance inference on Intel XPUs, CPUs, and NVIDIA GPUs.

Who it’s for

It is built for developers and engineers who need to deploy vision AI in production environments—such as factory floors for defect detection, medical imaging, or automated inventory management—without needing a massive data center or Kubernetes cluster.

Highlights

  • Comprehensive Model Catalog: Supports a wide array of tasks including object detection, instance segmentation, and image classification using modern architectures like YOLO26 and D-FINE.
  • Edge-Optimized: Built-in accuracy-aware INT8 quantization and OpenVINO export for low-latency deployment on resource-constrained hardware.
  • Smart Annotation: Reduces manual effort with AI-assisted bounding boxes and polygons.
  • Integrated Pipelines: Allows the creation of custom source-to-sink pipelines to monitor real-time predictions on video streams.

Related

  • Project
    open-edge-platform/geti_v2

    An end-to-end computer vision platform that uses active learning and smart annotations to build and deploy optimized AI models with minimal data.

  • Project
    nndeploy/nndeploy

    A high-performance AI deployment framework that uses visual workflows to deploy AI models across desktop, mobile, and edge devices.

  • Project
    openvinotoolkit/openvino

    An open-source toolkit for optimizing and deploying deep learning models across CPU, GPU, and NPU hardware to boost inference performance.

  • Project
    lightly-ai/lightly-train

    A computer vision framework for pretraining, fine-tuning, and distilling SOTA models like DINOv3 and LTDETRv2 for tasks such as object detection and segmentation.

  • Project
    ultralytics/ultralytics

    A high-performance computer vision framework implementing YOLO models for real-time object detection, segmentation, classification, and pose estimation.