MinIO | The Prompt API

The Prompt API

The promptObject API is a powerful new extension of the S3 API that lets users or applications talk to unstructured objects as if they were talking to an LLM - with zero domain knowledge required. That means you can ask an object to describe itself, to find similarities with other objects, to find differences with other objects.

promptObject API

Unstructured Objects

The promptObject API works with any unstructured object(s). Images, PDFs, GIFs, documents are all supported with video and audio optimizations in the near term roadmap.

Single or Multiple Objects

The promptObject API supports interaction with one or more objects. This enables the user or application to discern similarities, differences or compare in other ways.

Support for Function Calling

Function calling with additional logic allows conditional or context-specific logic to be added to adapt the promptObject API’s behavior based on certain parameters, user inputs, or external factors. This allows a function to perform different actions depending on its input or environment.

Support for Chained Functions

Chained functions enable the output of one promptObject API call to be passed as input to the next. This chaining allows for complex workflows by combining simple, modular functions, each of which performs a specific task. This allows users to perform multi-step operations on data objects within a single interaction.

API, GUI or SDK

The promptObject API can be consumed in the Global Console, a Jupyter Notebook or through the SDK. Choose your weapon!

Transparent but not Invisible

The underlying multi-modal LLM used by MinIO is transparent to the user. They need no domain specific expertise in RAG, vector databases or other AI specific knowledge. Having said that, the LLM is open to the data science teams for inspection.