* a model registry (so your code can be independent from your model weights & you don't have to re-build and re-deploy every time you want to iterate on your model weights) * reusable utils to make adding support for new models easier * foundation model implementations (like CLIP & SAM) that tend to chain well with fine-tuned models

* model architecture implementations (which implement the tensor parsing glue between images & predictions) for supervised models that you've fine-tuned to perform custom tasks * standardized APIs for computer vision tasks (so switching out the model weights and architecture can be done independently of your application code) * a server (so you don’t have to reimplement things like image processing & prediction visualization on every project) It’s backed by a bunch of component pieces: The end result is a Docker container that serves a standardized API as a microservice that your application uses to get predictions from computer vision models (though there is also a native Python interface).

It’s an easy to use inference server for computer vision models. Project mention: Show HN: Pip install inference, open source computer vision deployment | |