The official Python library for the anthropic API
The Anthropic Python library provides convenient access to the Anthropic 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.
For the AWS Bedrock API, see anthropic-bedrock
.
The API documentation can be found here.
pip install anthropic
The full API of this library can be found in api.md.
from anthropic import Anthropic, HUMAN_PROMPT, AI_PROMPT
anthropic = Anthropic(
# defaults to os.environ.get("ANTHROPIC_API_KEY")
api_key="my api key",
)
completion = anthropic.completions.create(
model="claude-2.1",
max_tokens_to_sample=300,
prompt=f"{HUMAN_PROMPT} how does a court case get to the Supreme Court?{AI_PROMPT}",
)
print(completion.completion)
While you can provide an api_key
keyword argument,
we recommend using python-dotenv
to add ANTHROPIC_API_KEY="my-anthropic-api-key"
to your .env
file
so that your API Key is not stored in source control.
Simply import AsyncAnthropic
instead of Anthropic
and use await
with each API call:
from anthropic import AsyncAnthropic, HUMAN_PROMPT, AI_PROMPT
anthropic = AsyncAnthropic(
# defaults to os.environ.get("ANTHROPIC_API_KEY")
api_key="my api key",
)
async def main():
completion = await anthropic.completions.create(
model="claude-2.1",
max_tokens_to_sample=300,
prompt=f"{HUMAN_PROMPT} how does a court case get to the Supreme Court?{AI_PROMPT}",
)
print(completion.completion)
asyncio.run(main())
Functionality between the synchronous and asynchronous clients is otherwise identical.
We provide support for streaming responses using Server Side Events (SSE).
from anthropic import Anthropic, HUMAN_PROMPT, AI_PROMPT
anthropic = Anthropic()
stream = anthropic.completions.create(
prompt=f"{HUMAN_PROMPT} Your prompt here{AI_PROMPT}",
max_tokens_to_sample=300,
model="claude-2.1",
stream=True,
)
for completion in stream:
print(completion.completion, end="", flush=True)
The async client uses the exact same interface.
from anthropic import AsyncAnthropic, HUMAN_PROMPT, AI_PROMPT
anthropic = AsyncAnthropic()
stream = await anthropic.completions.create(
prompt=f"{HUMAN_PROMPT} Your prompt here{AI_PROMPT}",
max_tokens_to_sample=300,
model="claude-2.1",
stream=True,
)
async for completion in stream:
print(completion.completion, end="", flush=True)
This library provides several conveniences for streaming messages, for example:
import asyncio
from anthropic import AsyncAnthropic
client = AsyncAnthropic()
async def main() -> None:
async with client.beta.messages.stream(
max_tokens=1024,
messages=[
{
"role": "user",
"content": "Say hello there!",
}
],
model="claude-2.1",
) as stream:
async for text in stream.text_stream:
print(text, end="", flush=True)
print()
message = await stream.get_final_message()
print(message.model_dump_json(indent=2))
asyncio.run(main())
Streaming with client.beta.messages.stream(...)
exposes various helpers for your convenience including event handlers and accumulation.
Alternatively, you can use client.beta.messages.create(..., stream=True)
which only returns an async iterable of the events in the stream and thus uses less memory (it does not build up a final message object for you).
You can estimate billing for a given request with the client.count_tokens()
method, eg:
client = Anthropic()
client.count_tokens('Hello world!') # 3
Nested request parameters are TypedDicts. Responses are Pydantic models, which provide helper methods for things like:
model.model_dump_json(indent=2, exclude_unset=True)
model.model_dump(exclude_unset=True)
Typed requests and responses 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
.
When the library is unable to connect to the API (for example, due to network connection problems or a timeout), a subclass of anthropic.APIConnectionError
is raised.
When the API returns a non-success status code (that is, 4xx or 5xx
response), a subclass of anthropic.APIStatusError
is raised, containing status_code
and response
properties.
All errors inherit from anthropic.APIError
.
import anthropic
client = anthropic.Anthropic()
try:
client.completions.create(
prompt=f"{anthropic.HUMAN_PROMPT} Your prompt here{anthropic.AI_PROMPT}",
max_tokens_to_sample=300,
model="claude-2.1",
)
except anthropic.APIConnectionError as e:
print("The server could not be reached")
print(e.__cause__) # an underlying Exception, likely raised within httpx.
except anthropic.RateLimitError as e:
print("A 429 status code was received; we should back off a bit.")
except anthropic.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 |
Certain errors are 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 are all retried by default.
You can use the max_retries
option to configure or disable retry settings:
from anthropic import Anthropic, HUMAN_PROMPT, AI_PROMPT
# Configure the default for all requests:
anthropic = Anthropic(
# default is 2
max_retries=0,
)
# Or, configure per-request:
anthropic.with_options(max_retries=5).completions.create(
prompt=f"{HUMAN_PROMPT} Can you help me effectively ask for a raise at work?{AI_PROMPT}",
max_tokens_to_sample=300,
model="claude-2.1",
)
By default requests time out after 10 minutes. You can configure this with a timeout
option,
which accepts a float or an httpx.Timeout
object:
from anthropic import Anthropic, HUMAN_PROMPT, AI_PROMPT
# Configure the default for all requests:
anthropic = Anthropic(
# default is 10 minutes
timeout=20.0,
)
# More granular control:
anthropic = Anthropic(
timeout=httpx.Timeout(60.0, read=5.0, write=10.0, connect=2.0),
)
# Override per-request:
anthropic.with_options(timeout=5 * 1000).completions.create(
prompt=f"{HUMAN_PROMPT} Where can I get a good coffee in my neighbourhood?{AI_PROMPT}",
max_tokens_to_sample=300,
model="claude-2.1",
)
On timeout, an APITimeoutError
is thrown.
Note that requests that time out are retried twice by default.
We automatically send the anthropic-version
header set to 2023-06-01
.
If you need to, you can override it by setting default headers per-request or on the client object.
Be aware that doing so may result in incorrect types and other unexpected or undefined behavior in the SDK.
from anthropic import Anthropic
client = Anthropic(
default_headers={"anthropic-version": "My-Custom-Value"},
)
We use the standard library logging
module.
You can enable logging by setting the environment variable ANTHROPIC_LOG
to debug
.
$ export ANTHROPIC_LOG=debug
None
means null
or missingIn 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}.')
The "raw" Response object can be accessed by prefixing .with_raw_response.
to any HTTP method call.
from anthropic import Anthropic, HUMAN_PROMPT, AI_PROMPT
anthropic = Anthropic()
response = anthropic.completions.with_raw_response.create(
model="claude-2.1",
max_tokens_to_sample=300,
prompt=f"{HUMAN_PROMPT} how does a court case get to the Supreme Court?{AI_PROMPT}",
)
print(response.headers.get('X-My-Header'))
completion = response.parse() # get the object that `completions.create()` would have returned
print(completion.completion)
These methods return an APIResponse
object.
You can directly override the httpx client to customize it for your use case, including:
import httpx
from anthropic import Anthropic
client = Anthropic(
# Or use the `ANTHROPIC_BASE_URL` env var
base_url="http://my.test.server.example.com:8083",
http_client=httpx.Client(
proxies="http://my.test.proxy.example.com",
transport=httpx.HTTPTransport(local_address="0.0.0.0"),
),
)
By default the library closes underlying HTTP connections whenever the client is garbage collected. You can manually close the client using the .close()
method if desired, or with a context manager that closes when exiting.
This package generally follows SemVer conventions, though certain backwards-incompatible changes may be released as minor versions:
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.
Python 3.7 or higher.