MetaApi is a powerful, fast, cost-efficient, easy to use and standards-driven cloud forex trading API for MetaTrader 4 and MetaTrader 5 platform designed for traders, investors and forex application developers to boost forex application development process. MetaApi can be used with any broker and does not require you to be a brokerage.
CopyFactory is a simple yet powerful copy-trading API which is a part of MetaApi. See below for CopyFactory readme section.
MetaApi is a paid service, but API access to one MetaTrader account is free of charge.
The MetaApi pricing was developed with the intent to make your charges less or equal to what you would have to pay for hosting your own infrastructure. This is possible because over time we managed to heavily optimize our MetaTrader infrastructure. And with MetaApi you can save significantly on application development and maintenance costs and time thanks to high-quality API, open-source SDKs and convenience of a cloud service.
Official REST and websocket API documentation: https://metaapi.cloud/docs/client/
This SDK requires a 3.8+ version of Python to run.
Please note that this SDK provides an abstraction over REST and websocket API to simplify your application logic.
For more information about SDK APIs please check docstring documentation in source codes located inside lib folder of this package.
Please check this short video to see how you can download samples via our web application.
You can find code examples at examples folder of our github repo or in the examples folder of the pip package.
We have composed a short guide explaining how to use the example code
pip install metaapi-cloud-sdkPlease use one of these ways:
- https://app.metaapi.cloud/token web UI to obtain your API token.
- An account access token which grants access to a single account. See section below on instructions on how to retrieve account access token.
Supply token to the MetaApi class constructor.
from metaapi_cloud_sdk import MetaApi
token = '...'
api = MetaApi(token=token)Account access token grants access to a single account. You can retrieve account access token via API:
account_id = '...'
account = await api.metatrader_account_api.get_account(account_id=account_id)
account_access_token = account.access_token
print(account_access_token)Alternatively, you can retrieve account access token via web UI on https://app.metaapi.cloud/accounts page (see this video).
Before you can use the API you have to add an MT account to MetaApi and start an API server for it.
You can manage MetaTrader accounts here: https://app.metaapi.cloud/accounts
To create an account, supply a request with account data and the platform field indicating the MetaTrader version. Provisioning profile id must not be included in the request for automatic broker settings detection.
try:
account = await api.metatrader_account_api.create_account(account={
'name': 'Trading account #1',
'type': 'cloud',
'login': '1234567',
'platform': 'mt4',
# password can be investor password for read-only access
'password': 'qwerty',
'server': 'ICMarketsSC-Demo',
'application': 'MetaApi',
'magic': 123456,
'quoteStreamingIntervalInSeconds': 2.5, # set to 0 to receive quote per tick
'reliability': 'regular' # set this field to 'high' value if you want to increase uptime of your account (recommended for production environments)
})
except Exception as err:
# process errors
if hasattr(err, 'details'):
# returned if the server file for the specified server name has not been found
# recommended to check the server name or create the account using a provisioning profile
if err.details == 'E_SRV_NOT_FOUND':
print(err)
# returned if the server has failed to connect to the broker using your credentials
# recommended to check your login and password
elif err.details == 'E_AUTH':
print(err)
# returned if the server has failed to detect the broker settings
# recommended to try again later or create the account using a provisioning profile
elif err.details == 'E_SERVER_TIMEZONE':
print(err)If the settings have not yet been detected for the broker, the server will begin the process of detection, and you will receive a response with wait time:
Retrying request in 60 seconds because request returned message: Automatic broker settings detection is in progress, please retry in 60 secondsThe client will automatically retry the request when the recommended time passes.
Several types of errors are possible during the request:
- Server file not found
- Authentication error
- Settings detection error
This error is returned if the server file for the specified server name has not been found. In case of this error it is recommended to check the server name. If the issue persists, it is recommended to create the account using a provisioning profile.
{
"id": 3,
"error": "ValidationError",
"message": "We were unable to retrieve the server file for this broker. Please check the server name or configure the provisioning profile manually.",
"details": "E_SRV_NOT_FOUND"
}This error is returned if the server has failed to connect to the broker using your credentials. In case of this error it is recommended to check your login and password, and try again.
{
"id": 3,
"error": "ValidationError",
"message": "We failed to authenticate to your broker using credentials provided. Please check that your MetaTrader login, password and server name are correct.",
"details": "E_AUTH"
}This error is returned if the server has failed to detect the broker settings. In case of this error it is recommended to retry the request later, or create the account using a provisioning profile.
{
"id": 3,
"error": "ValidationError",
"message": "We were not able to retrieve server settings using credentials provided. Please try again later or configure the provisioning profile manually.",
"details": "E_SERVER_TIMEZONE"
}If creating the account with automatic broker settings detection has failed, you can create it using a provisioning profile. To create an account using a provisioning profile, create a provisioning profile for the MetaTrader server, and then add the provisioningProfileId field to the request:
account = await api.metatrader_account_api.create_account(account={
'name': 'Trading account #1',
'type': 'cloud',
'login': '1234567',
# password can be investor password for read-only access
'password': 'qwerty',
'server': 'ICMarketsSC-Demo',
'provisioningProfileId': provisioningProfile.id,
'application': 'MetaApi',
'magic': 123456,
'quoteStreamingIntervalInSeconds': 2.5, # set to 0 to receive quote per tick
'reliability': 'regular' # set this field to 'high' value if you want to increase uptime of your account (recommended for production environments)
})# filter and paginate accounts, see doc for full list of filter options available
accounts = await api.metatrader_account_api.get_accounts(accounts_filter={
'limit': 10,
'offset': 0,
'query': 'ICMarketsSC-MT5',
'state': ['DEPLOYED']
})
# get accounts without filter (returns 1000 accounts max)
accounts = await api.metatrader_account_api.get_accounts();
account = await api.metatrader_account_api.get_account(account_id='accountId')await account.update(account={
'name': 'Trading account #1',
'login': '1234567',
# password can be investor password for read-only access
'password': 'qwerty',
'server': 'ICMarketsSC-Demo',
'quoteStreamingIntervalInSeconds': 2.5
})await account.remove()await account.deploy()
await account.undeploy()
await account.redeploy()Custom expert advisors can only be used for MT4 accounts on g1 infrastructure. EAs which use DLLs are not supported.
You can use the code below to create an EA. Please note that preset field is a base64-encoded preset file.
expert = await account.create_expert_advisor(expert_id='expertId', expert={
'period': '1h',
'symbol': 'EURUSD',
'preset': 'a2V5MT12YWx1ZTEKa2V5Mj12YWx1ZTIKa2V5Mz12YWx1ZTMKc3VwZXI9dHJ1ZQ'
})
await expert.upload_file('/path/to/custom-ea')experts = await account.get_expert_advisors()expert = await account.get_expert_advisor(expert_id='expertId')You can use the code below to update an EA. Please note that preset field is a base64-encoded preset file.
await expert.update(expert={
'period': '4h',
'symbol': 'EURUSD',
'preset': 'a2V5MT12YWx1ZTEKa2V5Mj12YWx1ZTIKa2V5Mz12YWx1ZTMKc3VwZXI9dHJ1ZQ'
})
await expert.upload_file('/path/to/custom-ea')await expert.remove()Provisioning profiles can be used as an alternative way to create MetaTrader accounts if the automatic broker settings detection has failed.
You can manage provisioning profiles here: https://app.metaapi.cloud/provisioning-profiles
# if you do not have created a provisioning profile for your broker,
# you should do it before creating an account
provisioningProfile = await api.provisioning_profile_api.create_provisioning_profile(profile={
'name': 'My profile',
'version': 5,
'brokerTimezone': 'EET',
'brokerDSTSwitchTimezone': 'EET'
})
# servers.dat file is required for MT5 profile and can be found inside
# config directory of your MetaTrader terminal data folder. It contains
# information about available broker servers
await provisioningProfile.upload_file(file_name='servers.dat', file='/path/to/servers.dat')
# for MT4, you should upload an .srv file instead
await provisioningProfile.upload_file(file_name='broker.srv', file='/path/to/broker.srv')provisioningProfiles = await api.provisioning_profile_api.get_provisioning_profiles()
provisioningProfile = await api.provisioning_profile_api.get_provisioning_profile(provisioning_profile_id='profileId')await provisioningProfile.update(profile={'name': 'New name'})
# for MT5, you should upload a servers.dat file
await provisioningProfile.upload_file(file_name='servers.dat', file='/path/to/servers.dat')
# for MT4, you should upload an .srv file instead
await provisioningProfile.upload_file(file_name='broker.srv', file='/path/to/broker.srv')await provisioningProfile.remove()RPC API let you query the trading terminal state. You should use RPC API if you develop trading monitoring apps like myfxbook or other simple trading apps.
connection = account.get_rpc_connection()
await connection.connect()
await connection.wait_synchronized()
# retrieve balance and equity
print(await connection.get_account_information())
# retrieve open positions
print(await connection.get_positions())
# retrieve a position by id
print(await connection.get_position(position_id='1234567'))
# retrieve pending orders
print(await connection.get_orders())
# retrieve a pending order by id
print(await connection.get_order(order_id='1234567'))
# retrieve history orders by ticket
print(await connection.get_history_orders_by_ticket(ticket='1234567'))
# retrieve history orders by position id
print(await connection.get_history_orders_by_position(position_id='1234567'))
# retrieve history orders by time range
print(await connection.get_history_orders_by_time_range(start_time=start_time, end_time=end_time))
# retrieve history deals by ticket
print(await connection.get_deals_by_ticket(ticket='1234567'))
# retrieve history deals by position id
print(await connection.get_deals_by_position(position_id='1234567'))
# retrieve history deals by time range
print(await connection.get_deals_by_time_range(start_time=start_time, end_time=end_time))connection = account.get_rpc_connection()
await connection.connect()
await connection.wait_synchronized()
# first, subscribe to market data
await connection.subscribe_to_market_data(symbol='GBPUSD')
# read symbols available
print(await connection.get_symbols())
# read contract specification
print(await connection.get_symbol_specification(symbol='GBPUSD'))
# read current price
print(await connection.get_symbol_price(symbol='GBPUSD'))
# unsubscribe from market data when no longer needed
await connection.unsubscribe_from_market_data(symbol='GBPUSD')Currently this API is supported on G1 only.
from datetime import datetime
# retrieve 1000 candles before the specified time
candles = await account.get_historical_candles(symbol='EURUSD', timeframe='1m',
start_time=datetime.fromisoformat('2021-05-01'), limit=1000)
# retrieve 1000 ticks after the specified time
ticks = await account.get_historical_ticks(symbol='EURUSD', start_time=datetime.fromisoformat('2021-05-01'),
offset=5, limit=1000)
# retrieve 1000 latest ticks
ticks = await account.get_historical_ticks(symbol='EURUSD', start_time=None, offset=0, limit=1000)Real-time streaming API is good for developing trading applications like trade copiers or automated trading strategies. The API synchronizes the terminal state locally so that you can query local copy of the terminal state really fast.
from datetime import datetime
account = await api.metatrader_account_api.get_account(account_id='accountId')
connection = account.get_streaming_connection()
await connection.connect()
# access local copy of terminal state
terminalState = connection.terminal_state
# wait until synchronization completed
await connection.wait_synchronized()
print(terminalState.connected)
print(terminalState.connected_to_broker)
print(terminalState.account_information)
print(terminalState.positions)
print(terminalState.orders)
# symbol specifications
print(terminalState.specifications)
print(terminalState.specification(symbol='EURUSD'))
print(terminalState.price(symbol='EURUSD'))
# access history storage
historyStorage = connection.history_storage
# both orderSynchronizationFinished and dealSynchronizationFinished
# should be true once history synchronization have finished
print(historyStorage.order_synchronization_finished)
print(historyStorage.deal_synchronization_finished)
print(historyStorage.deals)
print(historyStorage.get_deals_by_ticket('1'))
print(historyStorage.get_deals_by_position('1'))
print(historyStorage.get_deals_by_time_range(
datetime.fromtimestamp(datetime.now().timestamp() - 24 * 60 * 60), datetime.now())
print(historyStorage.history_orders)
print(historyStorage.get_history_orders_by_ticket('1'))
print(historyStorage.get_history_orders_by_position('1'))
print(historyStorage.get_history_orders_by_time_range(
datetime.fromtimestamp(datetime.now().timestamp() - 24 * 60 * 60), datetime.now())By default history is stored in memory only. You can override history storage to save trade history to a persistent storage like MongoDB database.
from metaapi_cloud_sdk import HistoryStorage
class MongodbHistoryStorage(HistoryStorage):
# implement the abstract methods, see MemoryHistoryStorage for sample
# implementation
historyStorage = MongodbHistoryStorage()
# Note: if you will not specify history storage, then in-memory storage
# will be used (instance of MemoryHistoryStorage)
connection = account.get_streaming_connection(history_storage=historyStorage)
await connection.connect()
# access history storage
historyStorage = connection.history_storage;
# invoke other methods provided by your history storage implementation
print(await historyStorage.yourMethod())You can override SynchronizationListener in order to receive synchronization event notifications, such as account/position/order/history updates or symbol quote updates.
from metaapi_cloud_sdk import SynchronizationListener
# receive synchronization event notifications
# first, implement your listener
class MySynchronizationListener(SynchronizationListener):
# override abstract methods you want to receive notifications for
# retrieving a connection
connection = account.get_streaming_connection(history_storage=historyStorage)
# now add the listener
listener = MySynchronizationListener()
connection.add_synchronization_listener(listener=listener)
# remove the listener when no longer needed
connection.remove_synchronization_listener(listener=listener)
# open the connection after adding listeners
await connection.connect()connection = account.get_streaming_connection()
await connection.connect()
await connection.wait_synchronized()
# first, subscribe to market data
await connection.subscribe_to_market_data(symbol='GBPUSD')
# read contract specification
print(terminalState.specification(symbol='EURUSD'))
# read current price
print(terminalState.price(symbol='EURUSD'))
# unsubscribe from market data when no longer needed
await connection.unsubscribe_from_market_data(symbol='GBPUSD')connection = account.get_rpc_connection()
# or
connection = account.get_streaming_connection()
await connection.connect()
await connection.wait_synchronized()
# trade
print(await connection.create_market_buy_order(symbol='GBPUSD', volume=0.07, stop_loss=0.9, take_profit=2.0,
options={'comment': 'comment', 'clientId': 'TE_GBPUSD_7hyINWqAl'}))
print(await connection.create_market_sell_order(symbol='GBPUSD', volume=0.07, stop_loss=2.0, take_profit=0.9,
options={'comment': 'comment', 'clientId': 'TE_GBPUSD_7hyINWqAl'}))
print(await connection.create_limit_buy_order(symbol='GBPUSD', volume=0.07, open_price=1.0, stop_loss=0.9,
take_profit=2.0, options={'comment': 'comment', 'clientId': 'TE_GBPUSD_7hyINWqAl'}))
print(await connection.create_limit_sell_order(symbol='GBPUSD', volume=0.07, open_price=1.5, stop_loss=2.0,
take_profit=0.9, options={'comment': 'comment', 'clientId': 'TE_GBPUSD_7hyINWqAl'}))
print(await connection.create_stop_buy_order(symbol='GBPUSD', volume=0.07, open_price=1.5, stop_loss=2.0,
take_profit=0.9, options={'comment': 'comment', 'clientId': 'TE_GBPUSD_7hyINWqAl'}))
print(await connection.create_stop_sell_order(symbol='GBPUSD', volume=0.07, open_price=1.0, stop_loss=2.0,
take_profit=0.9, options={'comment': 'comment', 'clientId': 'TE_GBPUSD_7hyINWqAl'}))
print(await connection.create_stop_limit_buy_order(symbol='GBPUSD', volume=0.07, open_price=1.5,
stop_limit_price=1.4, stop_loss=0.9, take_profit=2.0, options={'comment': 'comment',
'clientId': 'TE_GBPUSD_7hyINWqAl'}))
print(await connection.create_stop_limit_sell_order(symbol='GBPUSD', volume=0.07, open_price=1.0,
stop_limit_price=1.1, stop_loss=2.0, take_profit=0.9, options={'comment': 'comment',
'clientId': 'TE_GBPUSD_7hyINWqAl'}))
print(await connection.modify_position(position_id='46870472', stop_loss=2.0, take_profit=0.9))
print(await connection.close_position_partially(position_id='46870472', volume=0.9))
print(await connection.close_position(position_id='46870472'))
print(await connection.close_by(position_id='46870472', opposite_position_id='46870482'))
print(await connection.close_positions_by_symbol(symbol='EURUSD'))
print(await connection.modify_order(order_id='46870472', open_price=1.0, stop_loss=2.0, take_profit=0.9))
print(await connection.cancel_order(order_id='46870472'))
# if you need to, check the extra result information in stringCode and numericCode properties of the response
result = await connection.create_market_buy_order(symbol='GBPUSD', volume=0.07, stop_loss=0.9, take_profit=2.0,
options={'comment': 'comment', 'clientId': 'TE_GBPUSD_7hyINWqAl'}))
print('Trade successful, result code is ' + result['stringCode'])
# catch and output exception
try:
await connection.create_market_buy_order(symbol='GBPUSD', volume=0.07, stop_loss=0.9, take_profit=2.0,
options={'comment': 'comment', 'clientId': 'TE_GBPUSD_7hyINWqAl'})
except Exception as err:
print(api.format_error(err))Trailing stop loss is a trade option that allows you to automatically configure and change the order/position stop loss based on the current price of the symbol. The specified settings are run on the server and modify the stop loss regardless of your connection to the account. The stop loss can be modified no more often than once in 15 seconds. Two types of trailing stop loss are available: distance stop loss and threshold stop loss, but both can be specified at the same time. You can find the full description here: https://metaapi.cloud/docs/client/models/trailingStopLoss/
# distance trailing stop loss
print(await connection.create_market_buy_order('GBPUSD', 0.07, 0.9, 2.0, {
'trailingStopLoss': {
'distance': {
'distance': 200,
'units': 'RELATIVE_POINTS'
}
}
}))
# threshold trailing stop loss
print(await connection.create_market_buy_order('GBPUSD', 0.07, 0.9, 2.0, {
'trailingStopLoss': {
'thresholds': [
{
'threshold": 50,
'stopLoss": 100
},
{
'threshold": 100,
'stopLoss": 50
}
],
'units': 'RELATIVE_POINTS'
}
}))You can monitor account connection health using MetaApiConnection.health_monitor API.
monitor = connection.health_monitor
# retrieve server-side app health status
print(monitor.server_health_status)
# retrieve detailed connection health status
print(monitor.health_status)
# retrieve account connection update measured over last 7 days
print(monitor.uptime)You can track latencies using MetaApi.latency_monitor API. Client-side latencies include network communication delays, thus the lowest client-side latencies are achieved if you host your app in AWS Ohio region.
api = MetaApi('token', {'enableLatencyMonitor': True})
monitor = api.latency_monitor
# retrieve trade latency stats
print(monitor.trade_latencies)
# retrieve update streaming latency stats
print(monitor.update_latencies)
# retrieve quote streaming latency stats
print(monitor.price_latencies)
# retrieve request latency stats
print(monitor.request_latencies)Please note that not all MT4/MT5 servers allows you to create demo accounts using the method below.
demo_account = await api.metatrader_account_generator_api.create_mt4_demo_account(
account={
'balance': 100000,
'accountType': 'type',
'email': 'example@example.com',
'leverage': 100,
'serverName': 'Exness-Trial4',
'name': 'Test User',
'phone': '+12345678901'
})
# optionally specify a provisioning profile id if servers file is not found by server name
demo_account = await api.metatrader_account_generator_api.create_mt4_demo_account(
account={
'balance': 100000,
'accountType': 'type',
'email': 'example@example.com',
'leverage': 100,
'serverName': 'Exness-Trial4',
'name': 'Test User',
'phone': '+12345678901'
}, profile_id=provisioningProfile.id)live_account = await api.metatrader_account_generator_api.create_mt4_live_account(
account={
'balance': 100000,
'email': 'example@example.com',
'leverage': 100,
'serverName': 'Exness-Trial4',
'name': 'Test User',
'phone': '+12345678901',
'country': 'Unites States',
'zip': '12345',
'state': 'New York',
'city': 'New York',
'address': 'customer address'
})
# optionally specify a provisioning profile id if servers file is not found by server name
live_account = await api.metatrader_account_generator_api.create_mt4_live_account(
account={
'balance': 100000,
'email': 'example@example.com',
'leverage': 100,
'serverName': 'Exness-Trial4',
'name': 'Test User',
'phone': '+12345678901',
'country': 'Unites States',
'zip': '12345',
'state': 'New York',
'city': 'New York',
'address': 'customer address'
}, profile_id=provisioningProfile.id)demo_account = await api.metatrader_demo_account_api.create_mt5_demo_account(
account={
'accountType': 'type',
'balance': 100000,
'email': 'example@example.com',
'leverage': 100,
'serverName': 'ICMarketsSC-Demo'
})
# optionally specify a provisioning profile id if servers file is not found by server name
demo_account = await api.metatrader_account_generator_api.create_mt5_demo_account(
account={
'accountType': 'type',
'balance': 100000,
'email': 'example@example.com',
'leverage': 100,
'serverName': 'Exness-Trial4',
'name': 'Test User',
'phone': '+12345678901'
}, profile_id=provisioningProfile.id)live_account = await api.metatrader_account_generator_api.create_mt5_live_account(
account={
'accountType': 'type',
'balance': 100000,
'email': 'example@example.com',
'leverage': 100,
'serverName': 'Exness-Trial4',
'name': 'Test User',
'phone': '+12345678901',
'country': 'Unites States',
'zip': '12345',
'state': 'New York',
'city': 'New York',
'address': 'customer address'
})
# optionally specify a provisioning profile id if servers file is not found by server name
live_account = await api.metatrader_account_generator_api.create_mt5_live_account(
account={
'accountType': 'type',
'balance': 100000,
'email': 'example@example.com',
'leverage': 100,
'serverName': 'Exness-Trial4',
'name': 'Test User',
'phone': '+12345678901',
'country': 'Unites States',
'zip': '12345',
'state': 'New York',
'city': 'New York',
'address': 'customer address'
}, profile_id=provisioningProfile.id)By default SDK logs messages to console. You can select the SDK to use logging logging library by calling MetaApi.enable_logging() static method before creating MetaApi instances. .. code-block:: python
from metaapi_cloud_sdk import MetaApi
MetaApi.enable_logging()
meta_api = MetaApi(token)
Please note that the SDK does not configure logging automatically. If you decide to use logging, then your application is still responsible to configuring logging appenders and categories. Please refer to logging documentation for details.
API calls you make are subject to rate limits. See MT account management API and MetaApi API for details.
MetaApi applies quotas to the number of accounts and provisioning profiles, for more details see the MT account management API quotas
CopyFactory is a powerful trade copying API which makes developing forex trade copying applications as easy as writing few lines of code.
You can find CopyFactory Python SDK documentation here: https://github.com/agiliumtrade-ai/copyfactory-python-sdk
MetaStats is a powerful trade statistics API which makes it possible to add forex trading metrics into forex applications.
You can find MetaStats Python SDK documentation here: https://github.com/agiliumtrade-ai/metastats-python-sdk
MetaApi MT manager API is a cloud REST API which can be used to access and manage MT4 and MT5 servers.
You can find MT manager API documentation here: https://metaapi.cloud/docs/manager/
MetaApi risk management API is a cloud API for executing trading challenges and competitions. You can use this API for e.g. if you want to launch a proprietary trading company like FTMO. The API is also useful for trading firms/teams which have to enforce trading risk restrictions.
You can find MetaApi risk management Python SDK documentation here: https://github.com/agiliumtrade-ai/risk-management-python-sdk