What Are Embeddings in AI?
Embeddings are numerical representations (vectors) of text that capture semantic meaning. Similar concepts get similar vectors, enabling machines to understand that “dog” and “puppy” are related even though they share no letters.
How Embeddings Work
An embedding model converts text into a list of numbers (a vector), typically with 768 to 3072 dimensions:
"artificial intelligence" → [0.023, -0.841, 0.156, ..., 0.432]
"machine learning" → [0.019, -0.837, 0.149, ..., 0.445]
"chocolate cake" → [0.891, 0.234, -0.567, ..., -0.123]The first two vectors are close together (similar meaning); the third is far away.
What Embeddings Enable
- Semantic search: Find documents by meaning, not just keywords
- RAG: Retrieve relevant context for LLM prompts
- Clustering: Group similar documents automatically
- Recommendations: “Users who liked X also liked Y”
- Anomaly detection: Spot outliers in datasets
Popular Embedding Models
| Model | Dimensions | Provider |
|---|---|---|
| text-embedding-3-large | 3072 | OpenAI |
| Cohere Embed v3 | 1024 | Cohere |
| BGE-large | 1024 | BAAI (open-source) |
| Nomic Embed | 768 | Nomic (runs locally) |
Embeddings in Practice
Vector databases (Pinecone, Chroma, pgvector) store embeddings and enable fast similarity search — the backbone of any RAG system.
Elvean brings all these concepts together in one native Mac app — local models, cloud APIs, agentic tools, and more.
Learn more about Elvean