Client library for the openai API
Project description
OpenAI Python API Library
The OpenAI Python library provides convenient access to the OpenAI REST API from any Python 3.7+ application. It includes type definitions for all request params and response fields, and offers both synchronous and asynchronous clients powered by httpx.
Documentation
The API documentation can be found here.
Installation
pip install --pre openai
Usage
The full API of this library can be found in api.md.
from openai import OpenAI
client = OpenAI(
# defaults to os.environ.get("OPENAI_API_KEY")
api_key="my api key",
)
completion = client.chat.completions.create(
model="gpt-3.5-turbo",
messages=[
{
"role": "user",
"content": "Say this is a test",
}
],
)
print(completion.choices)
While you can provide an api_key keyword argument, we recommend using python-dotenv
and adding OPENAI_API_KEY="my api key" to your .env file so that your API Key is not stored in source control.
Async Usage
Simply import AsyncOpenAI instead of OpenAI and use await with each API call:
from openai import AsyncOpenAI
client = AsyncOpenAI(
# defaults to os.environ.get("OPENAI_API_KEY")
api_key="my api key",
)
async def main():
completion = await client.chat.completions.create(
model="gpt-3.5-turbo",
messages=[
{
"role": "user",
"content": "Say this is a test",
}
],
)
print(completion.choices)
asyncio.run(main())
Functionality between the synchronous and asynchronous clients is otherwise identical.
Streaming Responses
We provide support for streaming responses using Server Side Events (SSE).
from openai import OpenAI
client = OpenAI()
stream = client.completions.create(
prompt="Say this is a test",
model="text-davinci-003",
stream=True,
)
for part in stream:
print(part.choices[0].delta.content or "")
The async client uses the exact same interface.
from openai import AsyncOpenAI
client = AsyncOpenAI()
stream = await client.completions.create(
prompt="Say this is a test",
model="text-davinci-003",
stream=True,
)
async for part in stream:
print(part.choices[0].delta.content or "")
Module-level client
[!IMPORTANT]
We highly recommend instantiating client instances instead of relying on the global client.
We also expose a global client instance that is accessible in a similar fashion to versions prior to v1.
import openai
# optional; defaults to `os.environ['OPENAI_API_KEY']`
openai.api_key = '...'
# all client options can be configured just like the `OpenAI` instantiation counterpart
openai.base_url = "https://..."
openai.default_headers = {"x-foo": "true"}
completion = openai.chat.completions.create(
model="gpt-4",
messages=[
{
"role": "user",
"content": "How do I output all files in a directory using Python?",
},
],
)
print(completion.choices[0].message.content)
The API is the exact same as the standard client instance based API.
This is intended to be used within REPLs or notebooks for faster iteration, not in application code.
We recommend that you always instantiate a client (e.g., with client = OpenAI()) in application code because:
- It can be difficult to reason about where client options are configured
- It's not possible to change certain client options without potentially causing race conditions
- It's harder to mock for testing purposes
- It's not possible to control cleanup of network connections
Using Types
Nested request parameters are TypedDicts. Responses are Pydantic models, which provide helper methods for things like serializing back into json (v1, v2). To get a dictionary, you can call dict(model).
This helps provide autocomplete and documentation within your editor. If you would like to see type errors in VS Code to help catch bugs earlier, set python.analysis.typeCheckingMode to "basic".
Pagination
List methods in the OpenAI API are paginated.
This library provides auto-paginating iterators with each list response, so you do not have to request successive pages manually:
import openai
client = OpenAI()
all_jobs = []
# Automatically fetches more pages as needed.
for job in client.fine_tuning.jobs.list(
limit=20,
):
# Do something with job here
all_jobs.append(job)
print(all_jobs)
Or, asynchronously:
import asyncio
import openai
client = AsyncOpenAI()
async def main() -> None:
all_jobs = []
# Iterate through items across all pages, issuing requests as needed.
async for job in client.fine_tuning.jobs.list(
limit=20,
):
all_jobs.append(job)
print(all_jobs)
asyncio.run(main())
Alternatively, you can use the .has_next_page(), .next_page_info(), or .get_next_page() methods for more granular control working with pages:
first_page = await client.fine_tuning.jobs.list(
limit=20,
)
if first_page.has_next_page():
print(f"will fetch next page using these details: {first_page.next_page_info()}")
next_page = await first_page.get_next_page()
print(f"number of items we just fetched: {len(next_page.data)}")
# Remove `await` for non-async usage.
Or just work directly with the returned data:
first_page = await client.fine_tuning.jobs.list(
limit=20,
)
print(f"next page cursor: {first_page.after}") # => "next page cursor: ..."
for job in first_page.data:
print(job.id)
# Remove `await` for non-async usage.
Nested params
Nested parameters are dictionaries, typed using TypedDict, for example:
from openai import OpenAI
client = OpenAI()
client.files.list()
File Uploads
Request parameters that correspond to file uploads can be passed as bytes or a tuple of (filename, contents, media type).
from pathlib import Path
from openai import OpenAI
client = OpenAI()
contents = Path("input.jsonl").read_bytes()
client.files.create(
file=contents,
purpose="fine-tune",
)
The async client uses the exact same interface. This example uses aiofiles to asynchronously read the file contents but you can use whatever method you would like.
import aiofiles
from openai import OpenAI
client = OpenAI()
async with aiofiles.open("input.jsonl", mode="rb") as f:
contents = await f.read()
await client.files.create(
file=contents,
purpose="fine-tune",
)
Handling errors
When the library is unable to connect to the API (e.g., due to network connection problems or a timeout), a subclass of openai.APIConnectionError is raised.
When the API returns a non-success status code (i.e., 4xx or 5xx
response), a subclass of openai.APIStatusError will be raised, containing status_code and response properties.
All errors inherit from openai.APIError.
import openai
from openai import OpenAI
client = OpenAI()
try:
client.fine_tunes.create(
training_file="file-XGinujblHPwGLSztz8cPS8XY",
)
except openai.APIConnectionError as e:
print("The server could not be reached")
print(e.__cause__) # an underlying Exception, likely raised within httpx.
except openai.RateLimitError as e:
print("A 429 status code was received; we should back off a bit.")
except openai.APIStatusError as e:
print("Another non-200-range status code was received")
print(e.status_code)
print(e.response)
Error codes are as followed:
| Status Code | Error Type |
|---|---|
| 400 | BadRequestError |
| 401 | AuthenticationError |
| 403 | PermissionDeniedError |
| 404 | NotFoundError |
| 422 | UnprocessableEntityError |
| 429 | RateLimitError |
| >=500 | InternalServerError |
| N/A | APIConnectionError |
Retries
Certain errors will be automatically retried 2 times by default, with a short exponential backoff. Connection errors (for example, due to a network connectivity problem), 408 Request Timeout, 409 Conflict, 429 Rate Limit, and >=500 Internal errors will all be retried by default.
You can use the max_retries option to configure or disable this:
from openai import OpenAI
# Configure the default for all requests:
client = OpenAI(
# default is 2
max_retries=0,
)
# Or, configure per-request:
client.with_options(max_retries=5).chat.completions.create(
model="gpt-3.5-turbo",
messages=[
{
"role": "user",
"content": "How can I get the name of the current day in Node.js?",
}
],
)
Timeouts
Requests time out after 10 minutes by default. You can configure this with a timeout option,
which accepts a float or an httpx.Timeout:
from openai import OpenAI
# Configure the default for all requests:
client = OpenAI(
# default is 60s
timeout=20.0,
)
# More granular control:
client = OpenAI(
timeout=httpx.Timeout(60.0, read=5.0, write=10.0, connect=2.0),
)
# Override per-request:
client.with_options(timeout=5 * 1000).chat.completions.create(
model="gpt-3.5-turbo",
messages=[
{
"role": "user",
"content": "How can I list all files in a directory using Python?",
}
],
)
On timeout, an APITimeoutError is thrown.
Note that requests which time out will be retried twice by default.
Advanced
How to tell whether None means null or missing
In an API response, a field may be explicitly null, or missing entirely; in either case, its value is None in this library. You can differentiate the two cases with .model_fields_set:
if response.my_field is None:
if 'my_field' not in response.model_fields_set:
print('Got json like {}, without a "my_field" key present at all.')
else:
print('Got json like {"my_field": null}.')
Configuring custom URLs, proxies, and transports
You can configure the following keyword arguments when instantiating the client:
import httpx
from openai import OpenAI
client = OpenAI(
# Use a custom base URL
base_url="http://my.test.server.example.com:8083",
proxies="http://my.test.proxy.example.com",
transport=httpx.HTTPTransport(local_address="0.0.0.0"),
)
See the httpx documentation for information about the proxies and transport keyword arguments.
Managing HTTP resources
By default we will close the underlying HTTP connections whenever the client is garbage collected is called but you can also manually close the client using the .close() method if desired, or with a context manager that closes when exiting.
Versioning
This package generally attempts to follow SemVer conventions, though certain backwards-incompatible changes may be released as minor versions:
- Changes that only affect static types, without breaking runtime behavior.
- Changes to library internals which are technically public but not intended or documented for external use. (Please open a GitHub issue to let us know if you are relying on such internals).
- Changes that we do not expect to impact the vast majority of users in practice.
We take backwards-compatibility seriously and work hard to ensure you can rely on a smooth upgrade experience.
We are keen for your feedback; please open an issue with questions, bugs, or suggestions.
Requirements
Python 3.7 or higher.
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
