Turning Text into Numbers

Embeddings are dense vector representations of text that capture semantic meaning. Two texts about the same topic will have similar embeddings, even if they use completely different words.

Available Models (2026)

const embedding = await openai.embeddings.create({
  model: "text-embedding-3-small",
  input: "How do I reset my password?",
  dimensions: 1024  // Optional: reduce dimensions for efficiency
});
// Returns: { embedding: [0.0023, -0.0091, 0.0154, ...] }

Dimension Reduction

Both models support native dimension reduction. You can request fewer dimensions (e.g., 256, 512, 1024) to save storage and improve search speed with minimal accuracy loss.

Use Cases

• Semantic Search: Find documents by meaning, not keywords
• RAG: Retrieve relevant context for LLM prompts
• Clustering: Group similar content automatically
• Anomaly Detection: Find outliers in text datasets
• Recommendations: "Users who liked X also liked Y"