General Note

Rely on external embeddings and vector stores. OpenAI APIs do not scale for retrieval.

Generate embeddings using the create but store them externally with vector indices.

Embeddings API

Create Embedding

import openai

client = OpenAI()

file = client.files.create()

# using file
res = client.embeddings.create(
  model="text-embedding-ada-002",
  file_id="file-abc123"
  encoding_format="float"
)

# using input
res = client.embeddings.create(
  model="text-embedding-ada-002",
  input="my name is Sam",
  encoding_format="float"
)

Response from calling the embedding create:

res = {
  "object": "list",
  "data": [
    {
      "object": "embedding",
      "embedding": [
        0.0023064255,
        -0.009327292,
        .... (1536 floats total for ada-002)
        -0.0028842222,
      ],
      "index": 0
    }
  ],
  "model": "text-embedding-ada-002",
  "usage": {
    "prompt_tokens": 8,
    "total_tokens": 8
  }
}

data_vector = res['data'][0]['embedding']

Available Models (2025)

text-embedding-3-small
text-embedding-3-large
text-embedding-ada-002 # deprecated

Vector Stores

OpenAI also has its own vector stores.

• No need to use this.
• Mostly used for Assistants API.

# vector create
vector_store = client.vector_stores.create(
  name="Support FAQ",
)

client.vector_stores.files.upload_and_poll(
    vector_store_id=vector_store.id,
    file=open("customer_policies.txt", "rb")
)