Nutanix Enterprise AI (formerly GPT-in-a-Box) provides a specific set of OpenAI-compatible API endpoints to enable developers to interact with deployed models using familiar tools. The officially supported endpoints include /v1/models for listing available models, /v1/chat/completions for conversational inference, and /v1/embeddings for generating text embeddings. The endpoint /v1/systems is not a standard or supported OpenAI-compatible endpoint within the Nutanix Enterprise AI framework. Its inclusion in a list would be incorrect, as it is not part of the documented API surface.

B. /v1/embeddings: This is a supported endpoint used to create vector representations (embeddings) of text inputs for various machine learning tasks.

C. /v1/chat/completions: This is a core supported endpoint for interacting with language models in a conversational, turn-by-turn manner.

D. /v1/models: This is a supported endpoint that allows clients to retrieve a list of the large language models available for use.

1. Nutanix. (2024). Nutanix GPT-in-a-Box 2.0 Deployment and User Guide. Section: "Using the OpenAI-Compatible Endpoints", Page 31. This section explicitly lists the three supported endpoints: /v1/models, /v1/chat/completions, and /v1/embeddings. The /v1/systems endpoint is not mentioned.