TweetAPI Python SDK

tweetapi is the Python SDK for TweetAPI. It provides methods for users, tweets, posts, profiles, interactions, lists, communities, Spaces, search, authentication, and direct messages.

Install

pip install tweetapi

Quick start

from tweetapi import TweetAPI

client = TweetAPI(api_key="YOUR_API_KEY")

# Get a user profile
user = client.user.get_by_username(username="elonmusk")
print(user["data"]["followerCount"])

# Search tweets
results = client.explore.search(query="bitcoin", type="Latest")

# Get followers with pagination
followers = client.user.get_followers(user_id="123456")
next_page = client.user.get_followers(
    user_id="123456",
    cursor=followers["pagination"]["nextCursor"],
)

Create an API key with 100 included requests. No credit card is required.

SDK behavior

• Public methods have type annotations, and response models use TypedDict.
• The client retries rate limits, server errors, timeouts, and connection failures.
• paginate() yields items from cursor-based responses; paginate_pages() yields full response pages.
• HTTP errors map to exception classes such as RateLimitError and NotFoundError.
• Timeouts can use one value or separate connect and read values.
• The package supports Python 3.9+ and depends on requests.

API reference

User

Tweet

Post

Profile

Interaction

List

Community

Space

Explore

Auth

country is the ISO 3166-1 alpha-2 code for the proxy's public egress IP (for example, "US"). It must match the IP used for the complete login attempt. Pass two_factor_secret when the account uses TOTP-based 2FA.

X Chat (encrypted DMs)

Unencrypted DMs

Posting and profile media

Tweet media accepts an existing TweetAPI media ID, a URL, or inline base64 data:

client.post.create_post_with_media(
    auth_token="AUTH_TOKEN",
    text="Launch update",
    media=[{"media_id": "1971008286821380096"}],
    proxy="host:port:user:pass",
)

Use reply_option to set reply controls when creating standard, quote, or media posts:

client.post.create_post(
    auth_token="AUTH_TOKEN",
    text="Followers can reply",
    proxy="host:port:user:pass",
    reply_option={"mode": "followers"},
)

Profile avatar and banner uploads accept a URL or inline data plus type:

client.profile.update(auth_token="AUTH_TOKEN", name="New Name", bio="Builder")
client.profile.avatar(
    auth_token="AUTH_TOKEN",
    media={"url": "https://example.com/avatar.jpg"},
)
client.profile.banner(
    auth_token="AUTH_TOKEN",
    media={"data": "BASE64_IMAGE_DATA", "type": "image/png"},
)
client.profile.remove_banner(auth_token="AUTH_TOKEN")
client.profile.set_privacy(auth_token="AUTH_TOKEN", is_private=True)

Use client.list for list mutations. The older helpers under client.interaction remain available:

created = client.list.create(
    auth_token="AUTH_TOKEN",
    name="Research",
    description="Accounts to watch",
    is_private=True,
)
client.list.add_member(auth_token="AUTH_TOKEN", list_id=created["data"]["id"], user_id="123")
client.list.remove_member(auth_token="AUTH_TOKEN", list_id=created["data"]["id"], user_id="123")

Community quote posts support the same URL and media attachment shapes:

client.community.create_quote(
    auth_token="AUTH_TOKEN",
    text="Useful context",
    attachment_url="https://x.com/user/status/123",
    community_id="987",
    proxy="host:port:user:pass",
)
client.community.create_quote_with_media(
    auth_token="AUTH_TOKEN",
    text="Useful context",
    attachment_url="https://x.com/user/status/123",
    community_id="987",
    media=[{"media_id": "1971008286821380096"}],
    proxy="host:port:user:pass",
)

Pagination

paginate() yields individual items. paginate_pages() yields each full response page:

from tweetapi import TweetAPI, paginate, paginate_pages

client = TweetAPI(api_key="YOUR_API_KEY")

# Iterate individual items across all pages
for user in paginate(
    lambda cursor: client.user.get_followers(user_id="123456", cursor=cursor),
):
    print(user["username"])

# Iterate full pages (access page-level data)
for page in paginate_pages(
    lambda cursor: client.explore.search(query="bitcoin", type="Latest", cursor=cursor),
    max_pages=5,  # optional: limit number of pages
):
    print(f"Got {len(page['data'])} results")
    print(f"Next cursor: {page['pagination']['nextCursor']}")

Each helper accepts a callable that takes a cursor and returns a response with data and pagination.nextCursor. Use max_pages to limit the number of fetched pages.

Retries

The client retries these transient failures:

• For a 429 response, it waits for RateLimitError.retry_after seconds, up to max_retry_delay.
• For a 5xx response, timeout, or connection failure, it uses exponential backoff with up to 25% jitter.

By default, the client makes up to 3 retries. Its base delay starts at 1 second, doubles after each attempt, and is capped at 30 seconds. Other 4xx responses are not retried.

# Customize retry behavior
client = TweetAPI(
    api_key="YOUR_API_KEY",
    max_retries=5,             # default: 3
    initial_retry_delay=2.0,   # default: 1.0 (seconds)
    backoff_multiplier=3.0,    # default: 2.0
    max_retry_delay=60.0,      # default: 30.0 (seconds)
)

# Disable retries entirely
client = TweetAPI(api_key="YOUR_API_KEY", max_retries=0)

Rate-limit state

After a 429 response, client.rate_limit_info records the retry delay and the time the response was received:

print(client.rate_limit_info)

Error handling

The client raises retryable errors after it exhausts the configured retries. It raises other errors from the first response:

from tweetapi import (
    TweetAPI,
    TweetAPIError,
    AuthenticationError,
    RateLimitError,
    NotFoundError,
    ValidationError,
    ServerError,
    NetworkError,
)

client = TweetAPI(api_key="YOUR_API_KEY")

try:
    user = client.user.get_by_username(username="elonmusk")
except RateLimitError as e:
    print(f"Rate limited. Retry in {e.retry_after}s")
except NotFoundError:
    print("User not found")
except AuthenticationError:
    print("Invalid API key")
except ValidationError as e:
    print(f"Bad request: {e.message}")
except ServerError:
    print("API is having issues, try again later")
except NetworkError:
    print("Network error — check your connection")
except TweetAPIError as e:
    print(f"Error [{e.code}]: {e.message}")

Every error includes these attributes:

• code: API error code, such as "ACCOUNT_SUSPENDED" or "RATE_LIMIT"
• status_code: HTTP status code
• message: human-readable error message
• details: response context such as a field, reason, or retry delay

Configuration

client = TweetAPI(
    api_key="YOUR_API_KEY",           # Required
    base_url="https://...",           # Optional (default: https://api.tweetapi.com)
    timeout=30,                       # Optional; one value for connect and read
    connect_timeout=10.0,             # Optional (default: 10s)
    read_timeout=30.0,                # Optional (default: 30s)
    max_retries=3,                    # Optional (default: 3, set 0 to disable)
    backoff_multiplier=2.0,           # Optional (default: 2.0)
    initial_retry_delay=1.0,          # Optional (default: 1.0s)
    max_retry_delay=30.0,             # Optional (default: 30.0s)
)

# You can also pass a tuple for timeout
client = TweetAPI(api_key="YOUR_API_KEY", timeout=(5, 30))  # (connect, read)

Requirements

• Python 3.9+
• requests library

Links

• Documentation
• Create an API key
• Dashboard
• Node.js SDK

License

MIT

TweetAPI is a third-party service and is not affiliated with X Corp.