Telegram Channel Info Scraper

Resolve public Telegram channels, groups, bots and users by username — without a Telegram account, API key or browser.

This Actor parses the public t.me pages and returns one dataset row per requested username. It is useful for lead enrichment, channel vetting, ad-placement research and OSINT workflows.

What it returns

For each username:

• Entity type: channel, group, bot, user or unknown
• Title and public description
• confirmed: true when a titled entity page or channel preview was found; false for an unconfirmed contact page
• Telegram verification badge (verified)
• Subscriber / member / monthly-user count when publicly visible
• Avatar URL
• Optional avatar SHA-256 hash and Base64 data
• Per-row status: ok, not_found, invalid_input, rate_limited or error

Failed rows do not stop the run. The Actor keeps going and reports the error per username.

Input

{
  "usernames": ["telegram", "@durov", "https://t.me/examplebot"],
  "downloadAvatar": false,
  "maxConcurrency": 5,
  "requestTimeoutSecs": 20,
  "maxRetries": 2,
  "proxyConfiguration": {
    "useApifyProxy": false
  }
}

• usernames — required. Usernames, @mentions, t.me/... links and telegram.me/... links are accepted. t.me/s/... preview links are normalized to the username; private invite links are rejected.
• downloadAvatar — optional. Adds avatar SHA-256 and Base64 to successful rows. This triggers the optional avatar-downloaded charge.
• maxConcurrency — optional, 1–20. Default 5.
• requestTimeoutSecs — optional, 5–60. Default 20.
• maxRetries — optional, 0–5. Retries rate limits, timeouts and server errors. Default 2.
• proxyConfiguration — optional. Enable Apify Proxy if t.me blocks the platform IPs.

Output

Successful row:

{
  "input": "telegram",
  "username": "telegram",
  "type": "channel",
  "title": "Telegram News",
  "confirmed": true,
  "verified": true,
  "description": "The official Telegram on Telegram. Much recursion. Very Telegram. Wow.",
  "subscribers": 9174802,
  "avatar_url": "https://cdn4.telesco.pe/file/...jpg",
  "avatar_hash": null,
  "avatar_base64": null,
  "status": "ok",
  "http_status": 200,
  "error": null,
  "attempts": 1,
  "fetched_at": "2026-11-07T12:00:00+00:00",
  "updated_at": "2026-11-07T12:00:00+00:00"
}

Error row:

{
  "input": "nonexistent_abc12345",
  "username": "nonexistent_abc12345",
  "status": "not_found",
  "http_status": 404,
  "error": "Telegram entity not found",
  "attempts": 1,
  "fetched_at": "2026-11-07T12:00:00+00:00"
}

A run summary is stored in the default key-value store under the SUMMARY key.

Pricing setup

This Actor is designed for Apify's pay-per-event pricing. In Apify Console → Actor → Publication → Monetization, configure:

Event	Suggested price	When it is charged
entity-resolved	$0.001 per event ($1 / 1,000)	One charge per confirmed entity (titled page or recovered channel preview)
avatar-downloaded	$0.0005 per event ($0.50 / 1,000)	One charge per successfully downloaded avatar
apify-actor-start	keep the default	Actor start

Remove apify-default-dataset-item, or set its price to 0. Otherwise Apify will also charge the synthetic dataset event for error rows and may combine it with the explicit event.

Only confirmed resolutions are charged. not_found, invalid_input, other error rows, and unconfirmed contact pages (confirmed: false) are free.

You can change these prices at any time; Apify applies price increases after a 14-day notice period.

Limits and compliance

• Only public t.me pages are accessed. No login, no private groups, no member lists, no message history.
• Telegram may rate-limit or block cloud IP ranges. Lower maxConcurrency first; if that is not enough, enable Apify Proxy.
• Some usernames resolve to user pages without a public title. Those rows have type: "user", title: null, description: null and confirmed: false; they are free and are not 404s.
• t.me returns a contact page for any valid-looking username, so type: "user" with confirmed: false does not guarantee that the account exists. Channels are confirmed through the t.me/s/{username} preview.
• When t.me returns a degraded contact page for a channel, the Actor automatically falls back to the public t.me/s/{username} preview. Subscriber counts from that fallback are approximate (for example 9.17M). If the preview also fails, the row stays user or unknown; retry with lower maxConcurrency or enable Apify Proxy.
• HTML changes on Telegram can affect parsing. The Actor returns per-row errors instead of failing the whole batch.
• Data freshness depends on the public page at request time. Subscriber counts and descriptions are snapshots.
• This Actor is not affiliated with, endorsed by or sponsored by Telegram. Users are responsible for complying with Telegram's terms and applicable law.

Performance

• 1,000 mixed usernames: about 1–2 minutes.
• 5,000 channels: about 10 minutes; 512 MB memory is enough.
• Large batches need a longer run timeout. Use at least 1800 seconds, or split the input into smaller runs.
• When t.me rate-limits, the channel-preview fallback adds time but keeps channel results accurate.

Local development

Requires Python 3.12+ and uv.

uv sync
uv run pytest

Run the Actor locally with the Apify CLI:

mkdir -p storage/key_value_stores/default
cat > storage/key_value_stores/default/INPUT.json <<'JSON'
{
  "usernames": ["telegram", "durov"],
  "downloadAvatar": false
}
JSON

npx -y apify-cli@1.11.0 run

Deploy to Apify

This repository is connected to an Apify Actor. Pushes to main trigger a new build in Apify Console.

If you are not using the GitHub integration, you can still deploy manually:

npx -y apify-cli@1.11.0 login
npx -y apify-cli@1.11.0 push

Do not use both the GitHub integration and apify push for the same Actor — pick one source of truth.

中文说明

这是从 t-me-edge 的 Telegram 公开页面解析逻辑迁移出来的独立 Apify Actor。它不依赖原生产服务的 FastAPI、Redis 或部署环境，生产仓库不需要做任何改动。

• 输入用户名列表，按条输出实体信息；
• 只抓公开页面，不需要 Telegram 账号；
• 成功解析按 entity-resolved 事件计费，头像可选 avatar-downloaded；
• 失败行保留在数据集里，但不会中断整批任务。