Pipeline is an inference solution that abstracts away complexities of scalable, production machine learning deployment.
Simply submit a request to our REST API and you'll receive the output of the selected pipeline (a machine learning function or model) when the run completes.
Create/upload custom models using PipelineLibrary to run on our API or run models from our Model Hub, a selection of pre-trained models.
Interested in deploying your own models with our serverless API?
Checkout our custom models section from the docs!
This documentation outlines how to create custom models and how to use our pre-trained models on our serverless API.
- Access serverless GPU, this means no hourly rental fees, we charge on a pay-as-you-go basis. You can view our pricing here
- Up-to-date enterprise hardware. Our cloud is constantly being upgraded with Ampere and Volta GPUs.
- From a few requests to thousands, we handle the cloud infrastructure needed to run your ML models.
Updated 3 days ago