Embedding
A list of numbers (a vector, often 384 or 1536 long) that captures the meaning of a piece of text. Similar meanings get similar numbers.
Last updated
Analogy
A hash function that went to therapy: instead of scrambling everything, it puts similar ideas next to each other.
Code example
const res = await client.embeddings.create({ model: "text-embedding-3-small", input: "How do I revert a deploy?",}); const vector = res.data[0].embedding; // [0.12, -0.80, 0.33, ...]In an agent
Embeddings power semantic search, which is how agents find the right docs to read.
Practice
Flashcards
- What is an embedding?A list of numbers that represents the meaning of text. Similar meanings end up close together.
- Interview: four uses for embeddings?Semantic search, RAG, deduplication and clustering.
Interview questions
- What are embeddings?
- Vectors that represent meaning.
- Similar meanings give nearby vectors, measured with cosine similarity.
- They power semantic search, RAG, deduplication and clustering.
Questions people ask
What is an embedding in AI?
An embedding is a list of numbers, a vector often 384 or 1536 long, that captures the meaning of a piece of text. Texts with similar meanings get vectors that sit close together. That makes it possible to search, group or compare text by meaning instead of by exact words.
How do I create embeddings in JavaScript?
Call an embeddings endpoint, for example client.embeddings.create with model text-embedding-3-small and your text as input. The vector is in data[0].embedding. You can also run a small open model locally with Transformers.js. Store each vector next to its text, usually in a vector database.