Library / SDK
Inference & Serving
OpenVINO
1Intel 2toolkit 3for 4optimized 5model 6inference
the six Ws · specification
W1
Who
Developed and maintained by Intel.
W2
What
OpenVINO is a toolkit for optimizing and deploying deep learning models efficiently across Intel CPUs, GPUs and other hardware.
W3
Where
Hosted on GitHub at openvinotoolkit/openvino and distributed via PyPI and native installers.
W4
When
Released in 2018 and actively maintained.
W5
Why
It lets teams gain significant inference speedups on Intel hardware through graph optimization and quantization without changing model code.
W6
With
Converts and runs models exported from PyTorch, TensorFlow and ONNX.
W7
Watch
Apache 2.0 licensed, open source, maintained by Intel, roughly 10.6k GitHub stars.
works with