Retrieve Model

Get detailed information about a specific model

Overview

Retrieves detailed information about a specific model, including basic information such as the owner and permissioning details.

Method

client.models.retrieve(model="gpt-4-turbo")

Parameters

model · string · required

The model ID to retrieve. This can be any model identifier from the models list, including:

• Base models (e.g., gpt-4-turbo, gpt-3.5-turbo)
• Fine-tuned models (e.g., ft:gpt-3.5-turbo:org-id:model-name:id)

extra_headers · dict

Send extra headers with the request

extra_query · dict

Add additional query parameters to the request

timeout · float

Override the client-level default timeout for this request, in seconds

Response

Returns a Model object.

id · string

The model identifier that can be referenced in API endpoints

created · integer

Unix timestamp (in seconds) when the model was created

object · string

The object type, always “model”

owned_by · string

The organization that owns the model

Examples

Retrieve a Base Model

from openai import OpenAI

client = OpenAI()

model = client.models.retrieve("gpt-4-turbo")

print(f"Model ID: {model.id}")
print(f"Owner: {model.owned_by}")
print(f"Created: {model.created}")

Retrieve a Fine-Tuned Model

# Retrieve details about your fine-tuned model
model = client.models.retrieve(
    "ft:gpt-3.5-turbo:my-org:custom-model:id"
)

print(f"Fine-tuned model: {model.id}")
print(f"Owned by: {model.owned_by}")

Check Model Ownership

def check_model_ownership(model_id: str) -> str:
    try:
        model = client.models.retrieve(model_id)
        return model.owned_by
    except Exception as e:
        return f"Error: {e}"

owner = check_model_ownership("gpt-4")
print(f"Model owned by: {owner}")

Error Handling

try:
    model = client.models.retrieve("non-existent-model")
except Exception as e:
    print(f"Model not found: {e}")

Async Usage

from openai import AsyncOpenAI

client = AsyncOpenAI()

model = await client.models.retrieve("gpt-4-turbo")
print(model.id)

Notes

• Returns a 404 error if the model doesn’t exist or you don’t have access to it
• The model ID must be non-empty
• This endpoint is useful for verifying model details before use
• For fine-tuned models, you can only retrieve models owned by your organization