IBM watsonx.ai Runtime provides the resources that power the AI lifecycle for IBM watsonx.ai and Cloud Pak for Data as a Service. IBM watsonx.ai Runtime powers IBM watsonx.ai Studio. With resources for all AI solution tasks, you can build, test, and deploy machine learning models and on watsonx.ai, build, tune, and deploy generative AI solutions.

On watsonx.ai, inference foundation models with prompts that you customize for your solution. Build retrieval-augmented generation patterns that include extracting text, chunking and embedding documents for vector indexes, and reranking search results. Host the foundation models that you deploy and the endpoints for prompts and AI services.

Take advantage of machine learning models management (continuous learning system) and deployment (online, batch, streaming). Select any of widely supported machine learning frameworks: TensorFlow, Keras, Caffe, PyTorch, Spark MLlib, scikit learn, xgboost and SPSS.

Use the command line interface and Python client to manage your artifacts. Extend your application with artificial intelligence through the watsonx.ai REST API.

Create and train machine learning models with the best tools and the latest expertise in a social environment built by and for data scientists.