Features

• Unified AI Inference Support – Invoke chat completions, text completions, embeddings, and model discovery APIs through a single Volt MX Foundry data adapter.

• Dynamic Model Selection – Seamlessly switch between NVIDIA NIM hosted models using configurable model parameters.

• Context-Aware Conversations – Supports multi-turn chat interactions with message history for maintaining conversational context.

• Embeddings & Semantic AI Support – Generate embeddings for semantic search, retrieval-augmented generation (RAG), and similarity-based AI workflows.

• Configurable Response Controls – Customize AI output behavior using parameters such as temperature, top_p, and max_tokens.