Code beginner · 3 min read

How to use text-embedding-3-large in python

Q: How to use text-embedding-3-large in python

Use the OpenAI Python SDK's client.embeddings.create method with model="text-embedding-3-large" and your input text to generate embeddings.

Direct answer

Use the OpenAI Python SDK's client.embeddings.create method with model="text-embedding-3-large" and your input text to generate embeddings.

Setup

Install

bash

pip install openai

Env vars

OPENAI_API_KEY

Imports

python

import os
from openai import OpenAI

Examples

inThe quick brown fox jumps over the lazy dog.

out[0.0123, -0.0456, 0.0789, ...] # vector of floats representing the embedding

inOpenAI provides powerful AI models for developers.

out[0.0345, -0.0234, 0.0567, ...] # embedding vector output

out[] # empty input returns an empty or error response depending on API validation

Integration steps

Install the OpenAI Python SDK and set your OPENAI_API_KEY environment variable.
Import OpenAI from the openai package and initialize the client with your API key.
Call client.embeddings.create with model="text-embedding-3-large" and your input text as the input parameter.
Receive the response containing the embedding vector in response.data[0].embedding.
Use or store the embedding vector as needed for your application.

Full code

python

import os
from openai import OpenAI

# Initialize the OpenAI client with your API key from environment
client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

# Text to embed
text_to_embed = "The quick brown fox jumps over the lazy dog."

# Create embedding
response = client.embeddings.create(
    model="text-embedding-3-large",
    input=text_to_embed
)

# Extract the embedding vector
embedding_vector = response.data[0].embedding

print("Embedding vector length:", len(embedding_vector))
print("First 5 values:", embedding_vector[:5])

output

Embedding vector length: 1536
First 5 values: [0.012345, -0.045678, 0.078912, 0.003456, -0.009876]

API trace

Request

json

{"model": "text-embedding-3-large", "input": "The quick brown fox jumps over the lazy dog."}

Response

json

{"data": [{"embedding": [0.012345, -0.045678, 0.078912, ...]}], "usage": {"prompt_tokens": 10, "total_tokens": 10}}

Extractresponse.data[0].embedding

Variants

Batch Embedding Request ›

Use when you want to embed multiple texts in a single API call to reduce latency and cost.

python

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

texts = [
    "The quick brown fox jumps over the lazy dog.",
    "OpenAI provides powerful AI models for developers."
]

response = client.embeddings.create(
    model="text-embedding-3-large",
    input=texts
)

embeddings = [item.embedding for item in response.data]

for i, emb in enumerate(embeddings):
    print(f"Embedding {i} length: {len(emb)}")

Async Embedding Call ›

Use in asynchronous Python applications to avoid blocking while waiting for the embedding response.

python

import os
import asyncio
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

async def get_embedding(text: str):
    response = await client.embeddings.acreate(
        model="text-embedding-3-large",
        input=text
    )
    return response.data[0].embedding

async def main():
    embedding = await get_embedding("Async embedding example text.")
    print("Embedding length:", len(embedding))

asyncio.run(main())

Alternative Model: text-embedding-3-small ›

Use when you need faster embeddings with smaller vector size and lower cost, trading off some accuracy.

python

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

response = client.embeddings.create(
    model="text-embedding-3-small",
    input="A shorter, faster embedding model example."
)

embedding = response.data[0].embedding
print("Embedding length:", len(embedding))

Performance

Latency~300-600ms per embedding request for single input on text-embedding-3-large

Cost~$0.0004 per 1,000 tokens embedded with text-embedding-3-large

Rate limitsDefault tier: 60 RPM (requests per minute) and 60,000 TPM (tokens per minute)

Preprocess and truncate input text to relevant content to reduce token usage.
Batch multiple inputs in one request to amortize overhead.
Cache embeddings for repeated texts to avoid redundant calls.

Approach	Latency	Cost/call	Best for
Single input embedding	~300-600ms	~$0.0004	Embedding single texts with high accuracy
Batch embedding	~500-900ms	~$0.0004 per 1k tokens total	Embedding multiple texts efficiently
text-embedding-3-small model	~200-400ms	~$0.0002	Faster, cheaper embeddings with smaller vectors

✓

Quick tip

Always batch multiple texts in a single <code>embeddings.create</code> call to optimize throughput and reduce latency.

⚠

Common mistake

Passing an empty string or non-string input to <code>input</code> without validation, causing API errors or empty embeddings.

Verified 2026-04 · text-embedding-3-large, text-embedding-3-small

Verify ↗