Listing Available OpenAI Models – OpenAI API

OpenAI models are quite hot topic right now. OpenAI has a very useful API structure which you can access with Python and leverage this edge-cutting modern technology.

In this tutorial, we will demonstrate a few examples of OpenAI’s API using Python and the openai library.

Contents

Introduction
Listing OpenAI models through the API with Python
- openai Python Library
- OpenAI API Authorization
- openai.Model.list()
- Full Python code so far

How to serialize OpenAI API Data as Json
Summary

2) Listing OpenAI models through the API with Python

When I started using OpenAI’s API, I wanted to first list all of the models so I can be aware of different options out there. This was a great little exercise because you also get familiar with the authorization of your account for making API calls.

You can simply replace the string in the first line with your OpenAI API key.

`openai` Python Library

First step is importing the openai library which we will use to access OpenAI’s API. It can be imported simply as below.

import openai

If you’re missing the library in your system it can be installed using pip commands. Just type:

pip install openai

OpenAI API Authorization

After importing the openai library, you can authorize your session with OpenAI via the Python code below. We will use the api_key function as openai.api_key for that purpose. Check out 2 lines below:

OPENAI_API_KEY = 'YOUR_OPENAI_KEY'
openai.api_key = OPENAI_API_KEY

openai.Model.list()

Now that we imported the openai library and authorized our session with the API key, we can execute the actual function which will list all of the available OpenAI models for us.

For that purpose, we will use list function of Model class from the openai library as openai.Model.list().

model_lst = openai.Model.list()

This will return a big dictionary with two keys: “data” and “object”. If we look into the “data” key we will see all the models listed with various parameters.

If you are curious object key just shows the structure of the data information which is in this case “list” showing it’s a Python list. We will mainly explore “data” part of the returned object.

Below is 1st sample element from the data. Inside that list, we see a nested dictionary with various keys and another list as the value of the permission key.

The keys are:

created
id
object

owned_by
parent
permission

root

And permission key has its own nested list which has a further nested dictionary.

This nested data structure is quite common and wildly used in Rest API technologies combined with Json. To refresh your knowledge you can visit:

Rest API with Python

Let’s explore the values of the data’s first item.

{
“created”: 1649358449,
“id”: “babbage”,
“object”: “model”,
“owned_by”: “openai”,
“parent”: null,
“permission”: [
{
“allow_create_engine”: false,
“allow_fine_tuning”: false,
“allow_logprobs”: true,
“allow_sampling”: true,
“allow_search_indices”: false,
“allow_view”: true,
“created”: 1669085501,
“group”: null,
“id”: “modelperm-49FUp5v084tBB49tC4z8LPH5”,
“is_blocking”: false,
“object”: “model_permission”,
“organization”: “*”
}
],
“root”: “babbage”
},
{

As you can see there are quite a few technical parameters. However, “id” key gives the name of the model. We can use this information to list the names of all of the available models by OpenAI.

model_lst = openai.Model.list()

for i in model_lst['data']:
    print(i['id'])

This code will return all of the AI models that are available by OpenAI’s API including the popular gpt-3.5 turbo.

babbage
davinci
gpt-3.5-turbo-0301
text-davinci-003
babbage-code-search-code
text-similarity-babbage-001
text-davinci-001
ada
curie-instruct-beta
babbage-code-search-text
babbage-similarity
gpt-3.5-turbo
code-davinci-002
code-search-babbage-text-001
text-embedding-ada-002
code-cushman-001
whisper-1
code-search-babbage-code-001
audio-transcribe-deprecated
text-ada-001
text-similarity-ada-001
text-davinci-insert-002
ada-code-search-code
ada-similarity
code-search-ada-text-001
text-search-ada-query-001
text-curie-001
text-davinci-edit-001
davinci-search-document
ada-code-search-text
text-search-ada-doc-001
code-davinci-edit-001
davinci-instruct-beta
text-similarity-curie-001
code-search-ada-code-001
ada-search-query
text-search-davinci-query-001
curie-search-query
davinci-search-query
text-davinci-insert-001
babbage-search-document
ada-search-document
text-search-curie-query-001
text-search-babbage-doc-001
text-davinci-002
curie-search-document
text-search-curie-doc-001
babbage-search-query
text-babbage-001
text-search-davinci-doc-001
text-search-babbage-query-001
curie-similarity
curie
text-similarity-davinci-001
davinci-similarity
cushman:2020-05-03
ada:2020-05-03
babbage:2020-05-03
curie:2020-05-03
davinci:2020-05-03
if-davinci-v2
if-curie-v2
if-davinci:3.0.0
davinci-if:3.0.0
davinci-instruct-beta:2.0.0
text-ada:001
text-davinci:001
text-curie:001
text-babbage:001

Full Python code so far

import openai

OPENAI_API_KEY = 'YOUR_OPENAI_KEY'
openai.api_key = OPENAI_API_KEY


model_lst = openai.Model.list()

for i in model_lst['data']:
    print(i['id'])

I think another crucial step is to serialize this data, so we can save the data permanently and refer to it whenever needed instead of making separate API calls which will unnecessarily use resources and spend our API credits.

In the next session we will demonstrate how to save the returned object as a json file locally.

3) How to serialize OpenAI API Data as Json

You can save the OpenAI model list as json so you don’t have to make a request each time you need to refer to this data. This makes the information permanently written in your disk and it’s a technique called serialization.

Below, you can find a Python code that can be used to save our data in model_list.json file.

# Save OponAI model list as json

with open('model_list.json', 'w') as modelsjson:
    modelsjson.write(str(model_lst))

Finally, an important point to pay attention here is to make sure the object we are saving in the json file is a string. Originally, we mentioned that, OpenAI API returns a dictionary with two keys: data and object.

To save the result of the original call we are using str function to convert the outermost dictionary object to a string and save it to our json file. You can read more about this in our Data Type Conversions Python tutorial.

Summary

In this Python Tutorial we’ve learned how to access OpenAI’s powerful Rest API, how to authorize our API session and how to make calls to it. Particularly we demonstrated a Rest API call that returns a list of all of the available AI models of OpenAI.

We have also went over various fundamental Python topics such as nested data structures, data type conversion (in this case from dictionary to string), json structure and saving data as a json file,

If you don’t have the openai library for Python we’ve also discussed how it can be installed using pip package manager of Python. For further instructions on using pip, you can refer to: Installing pip packages with Python tutorial.