Intel Geti

Intel Geti

Build and run vision models locally

Description

Building an inspection or detection model means different tools for labeling, training, compression and deployment. Geti from Intel is an end-to-end platform for computer vision models from raw images to live inference, all local.

It ships as a Docker container or native Windows app guiding dataset prep, training, optimization and deployment, optimized for Intel XPU with CPU and CUDA support.

Features



End to end:Label, train, optimize and deploy.

Tasks:Classification, detection and segmentation.

Compression:Quantization and OpenVINO.

Local:Docker or Windows app.