Agentic AI Reference Architecture

Knowledge & Data · Model asset

Retrieval-Optimized Embedding Model

Model assetKnowledge & DataKnowledge & Dataarc:RetrievalOptimizedEmbeddingModel

A compact dense embedding model trained specifically for search ranking, excelling at distinguishing near-duplicate documents within short passages.

Responsibility. Encodes short passages into vectors optimized for retrieval precision and ranking.

Variant of Text Embedding Model abstract

When to choose. Choose for search-specific applications over short passages where ranking precision and near-duplicate discrimination matter more than long context.

deployed onspecializesis target of alternativeTois target of alternativeTois target of alternativeToText Embedding Service: deployed onText Embedding ServiceText Embedding Model: specializesText Embedding ModelHigh-Accuracy General Embedding Model: is target of alternativeToHigh-Accuracy General Em…Long-Context Embedding Model: is target of alternativeToLong-Context Embedding M…Cost-Optimized Embedding Model: is target of alternativeToCost-Optimized Embedding…
Direct neighbourhood (hover for relationship types)

Relationships

deployed on structural

alternative to variability

Quantitative guidance

As stated by the sources; verify before use.

Classification

Technologies
Cohere embed-english-v3Llama 3.2 EmbedQA 1B V2

Sources

  1. Ch6.1: T. Nguyen, "Embeddings and RAG Fundamentals," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 6.1. ISBN: 9798244538229.
  2. Ref7.07: E. Li, V. Bellotti, R. Kraus, and R. Kao, "Build a retrieval-augmented generation (RAG) agent with NVIDIA Nemotron," NVIDIA Technical Blog, Sep. 23, 2025. [Online]. Available: https://developer.nvidia.com/blog/build-a-rag-agent-with-nvidia-nemotron/