Telegram Scraper — Channels, Messages, Media & Search (No Login)

Extract data from public Telegram channels at scale — no login, no phone number, no API keys. Paste channel usernames or t.me links and get clean, structured JSON: full message history, media, reactions, polls, forwards, replies, link previews, and channel statistics. Includes in-channel keyword search, which most Telegram scrapers don't offer.

Built HTTP-only (no headless browser) for fast, low-cost, reliable runs.

What you can do

• Scrape full message history from any public channel (messages mode), oldest-to-newest, with no early page cap.
• Search inside a channel by keyword — uses Telegram's native channel search to return only matching messages.
• Pull channel metadata (channelInfo mode): title, username, description, avatar, subscriber/photo/video/file/link counts.
• Filter by date — only messages newer than / older than a given date (ISO or relative like 7 days, 3 months).
• Download media (photos, videos, voice, documents) into the run's key-value store. Media URLs are always included.

Use cases

• Crypto / web3 community & trading-signal monitoring
• OSINT, threat intelligence, and news aggregation
• Brand, topic, and competitor tracking
• Dataset building for LLM/RAG pipelines and sentiment analysis

Input

Field	Type	Description
channels (required)	array	Channel usernames or links: @telegram, telegram, https://t.me/telegram.
resultsType	string	messages (default) or channelInfo.
maxMessages	integer	Max messages per channel (default 100).
searchQuery	string	Optional keyword — returns only matching messages in each channel.
oldestMessageDate	string	Stop once messages are older than this (ISO or 7 days).
newestMessageDate	string	Skip messages newer than this.
downloadMedia	boolean	Download attached media into the key-value store (default false).
proxyConfiguration	object	Proxy settings (Apify Proxy datacenter recommended).

Example input

{
    "channels": ["@durov", "https://t.me/telegram"],
    "resultsType": "messages",
    "maxMessages": 200,
    "oldestMessageDate": "30 days",
    "downloadMedia": false
}

Output

Each message is one dataset record:

{
    "channel": "durov",
    "messageId": 372,
    "url": "https://t.me/durov/372",
    "date": "2026-05-21T14:02:11+00:00",
    "text": "Example post text...",
    "views": 1200000,
    "author": "Pavel Durov",
    "forwardedFrom": null,
    "replyToUrl": null,
    "isPinned": false,
    "hashtags": ["#telegram"],
    "mentions": ["@telegram"],
    "links": ["https://example.com"],
    "media": [{ "type": "photo", "url": "https://cdn5.telesco.pe/file/..." }],
    "poll": null,
    "linkPreview": null
}

In channelInfo mode each record contains title, username, description, avatarUrl, subscribers, and content counts.

Pricing (pay-per-event)

This actor charges per result, so you only pay for data you receive. Configure these event names in the Apify console:

Event	Suggested price
actor-start	$0.001
message	$0.002
channel-info	$0.003
media	$0.001

At ~$2 per 1,000 messages it undercuts existing per-post Telegram scrapers while returning far richer data.

Notes & limits

• Public channels only. Private/invite-only channels and the messages of standard user accounts are not accessible via the public web preview and are skipped with a clear log message.
• Reaction counts and some fields are included only when Telegram exposes them on the public preview.
• searchQuery performs within-channel search (per the channels you provide); it does not discover new channels across all of Telegram.

Run via API

The actor is fully runnable via the Apify API and integrates with Make, Zapier, n8n, and webhooks like any other actor.