Skip to main content

Alpaca API python client

Project description


alpaca-trade-api-python is a python library for the Alpaca trade API. It allows rapid trading algo development easily, with support for the both REST and streaming interfaces. For details of each API behavior, please see the online API document.

Note this module supports only python version 3.5 and above, due to the async/await keyword use.


$ pip3 install alpaca-trade-api


In order to call Alpaca's trade API, you need to obtain API key pairs. Replace <key_id> and <secret_key> with what you get from the web console.

REST example

import alpaca_trade_api as tradeapi

api = tradeapi.REST('<key_id>', '<secret_key>')
account = api.get_account()

API Document

The HTTP API document is located in


The Alpaca API requires API key ID and secret key, which you can obtain from the web console after you sign in. You can give them to the initializers of REST or StreamConn as arguments, or set up environment variables as follows.

  • APCA_API_SECRET_KEY: secret key

Base URL

The base URL for API calls defaults to This endpoint is for live trading, and for paper trading and other purposes, you can to change the base URL. You can pass it as an argument REST(), or using the environment variable, APCA_API_BASE_URL.


The REST class is the entry point for the API request. The instance of this class provides all REST API calls such as account, orders, positions, bars, quotes and fundamentals.

Each returned object is wrapped by a subclass of Entity class (or a list of it). This helper class provides property access (the "dot notation") to the json object, backed by the original object stored in the _raw field. It also converts certain types to the appropriate python object.

import alpaca_trade_api as tradeapi

api = tradeapi.REST()
account = api.get_account()

The Entity class also converts timestamp string field to a pandas.Timestamp object. Its _raw property returns the original raw primitive data unmarshaled from the response JSON text.

When a REST API call sees the 429 or 504 status code, this library retries 3 times by default, with 3 seconds apart between each call. These are configurable with the following environment variables.

  • APCA_RETRY_MAX: the number of subsequent API calls to retry, defaults to 3
  • APCA_RETRY_WAIT: seconds to wait between each call, defaults to 3
  • APCA_RETRY_CODES: comma-separated HTTP status code for which retry is attempted

If the retry exceeds, or other API error is returned, is raised. You can access the following information through this object.

  • the API error code: .code property
  • the API error message: str(error)
  • the original request object: .request property
  • the original response objecgt: .response property
  • the HTTP status code: .status_code property


Calls GET /account and returns an Account entity.


Calls GET /orders and returns a list of Order entities.

REST.submit_order(symbol, qty, side, type, time_in_force, limit_price=None, stop_price=None, client_order_id=None)

Calls POST /orders and returns an Order entity.


Calls GET /orders with client_order_id and returns an Order entity.


Calls GET /orders/{order_id} and returns an Order entity.


Calls DELETE /orders/{order_id}.


Calls GET /positions and returns a list of Position entities.


Calls GET /positions/{symbol} and returns a Position entity.

REST.list_assets(status=None, asset_class=None)

Calls GET /assets and returns a list of Asset entities.


Calls GET /assets/{symbol} and returns an Asset entity.


Calls GET /clock and returns a Clock entity.

REST.get_calendar(start=None, end=None)

Calls GET /calendar and returns a Calendar entity.


The StreamConn class provides WebSocket/NATS-based event-driven interfaces. Using the on decorator of the instance, you can define custom event handlers that are called when the pattern is matched on the channel name. Once event handlers are set up, call the run method which runs forever until a critical exception is raised. This module itself does not provide any threading capability, so if you need to consume the messages pushed from the server, you need to run it in a background thread.

This class provides a unique interface to the two interfaces, both Alpaca's account/trade updates events and Polygon's price updates. One connection is established when the subscribe() is called with the corresponding channel names. For example, if you subscribe to account_updates, a WebSocket connects to Alpaca stream API, and if AM.* given to the subscribe() method, a NATS connection is established to Polygon's interface.

The run method is a short-cut to start subscribing to channels and runnnig forever. The call will be blocked forever until a critical exception is raised, and each event handler is called asynchronously upon the message arrivals.

The run method tries to reconnect to the server in the event of connection failure. In this case you may want to reset your state which is best in the connect event. The method still raises exception in the case any other unknown error happens inside the event loop.

The msg object passed to each handler is wrapped by the entity helper class if the message is from the server.

Each event handler has to be a marked as async. Otherwise, a ValueError is raised when registering it as an event handler.

conn = StreamConn()

async def on_account_updates(conn, channel, account):
    print('account', account)

def on_bars(conn, channel, bar):
    print('bars', bar)

# blocks forever['account_updates', 'AM.*'])

You will likely call the run method in a thread since it will keep runnig unless an exception is raised.


Request "listen" to the server. channels must be a list of string channel names.

Goes into an infinite loop and awaits for messages from the server. You should set up event listeners using the on or register method before calling run.


As in the above example, this is a decorator method to add an event handler function. channel_pat is used as a regular expression pattern to filter stream names.

StreamConn.register(channel_pat, func)

Registers a function as an event handler that is triggered by the stream events that match with channel_path regular expression. Calling this method with the same channel_pat will overwrite the old handler.


Deregisters the event handler function that was previously registered via on or register method.

Support and Contribution

For technical issues particular to this module, please report the issue on this GitHub repository. Any API issues can be reported through Alpaca's customer support.

New features, as well as bug fixes, by sending pull request is always welcomed.

Polygon API Service

Alpaca's API key ID can be used to access Polygon API whose document is found here. This python SDK wraps their API service and seamlessly integrates with Alpaca API. alpaca_trade_api.REST.polygon will be the REST object for Polygon.

The example below gives AAPL daily OHLCV data in a DataFrame format.

import alpaca_trade_api as tradeapi

api = tradeapi.REST()
aapl = api.polygon.historic_agg('day', 'AAPL', limit=1000).df


It is initialized through alpaca REST object.


Returns a list of Exchange entity.


Returns a SymbolTypeMap object.

polygon/REST.historic_trades(symbol, date, offset=None, limit=None)

Returns a Trades which is a list of Trade entities.

  • date is a date string such as '2018-2-2'. The returned quotes are from this day onyl.
  • offset is an integer in Unix Epoch millisecond as the lower bound filter, inclusive.
  • limit is an integer for the number of ticks to return. Default and max is 30000.


Returns a pandas DataFrame object with the ticks returned by the historic_trades.

polygon/REST.historic_quotes(symbol, date, offset=None, limit=None)

Returns a Quotes which is a list of Quote entities.

  • date is a date string such as '2018-2-2'. The returned quotes are from this day only.
  • offset is an integer in Unix Epoch millisecond as the lower bound filter, inclusive.
  • limit is an integer for the number of ticks to return. Default and max is 30000.


Returns a pandas DataFrame object with the ticks returned by the historic_quotes.

polygon/REST.historic_agg(size, symbol, _from=None, to=None, limit=None)

Returns an Aggs which is a list of Agg entities. Aggs.df gives you the DataFrame object.

  • _from is an Eastern Time timestamp string that filters the result for the lower bound, inclusive.
  • to is an Eastern Time timestamp string that filters the result for the upper bound, inclusive.
  • limit is an integer to limit the number of results. 3000 is the default and max value.

Specify the _from parameter if you specify the to parameter since when to is specified _from is assumed to be the beginning of history. Otherwise, when you use only the limit or no parameters, the result is returned from the latest point.

The returned entities have fields relabeled with the longer name instead of shorter ones. For example, the o field is renamed to open.


Returns a pandas DataFrame object with the ticks returned by the hitoric_agg.


Returns a Trade entity representing the last trade for the symbol.


Returns a Quote entity representing the last quote for the symbol.


Returns a ConditionMap entity.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for alpaca-trade-api, version 0.18
Filename, size File type Python version Upload date Hashes
Filename, size alpaca_trade_api-0.18-py3-none-any.whl (16.0 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size alpaca-trade-api-0.18.tar.gz (13.0 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page