TikTok Scraper - Profiles, Videos, Hashtags & Search

⚡ TikTok Scraper — Profiles, Videos, Hashtags & Search (No Login, No API Key)

Scrape TikTok without a login or API key. Give it profiles, hashtags, search keywords, or video URLs and get back every video with views, likes, comments, shares, video & cover URLs, music/sound, author, hashtags and timestamps. Export to JSON, CSV or Excel, call it from the API, or have results delivered straight to Notion. Just $0.25 per 1,000 results — the cheapest TikTok scraper on the market, built for influencer discovery, trend tracking, lead gen and market research.

> No account. No signing keys. No msToken to paste. Paste a username and run.

Why this scraper wins

• 💸 Cheapest on the market — a flat $0.25 / 1,000 results. Billed per result actually returned, with no volume tiers and no plan gates. Everything you see in a row (stats, music, author, video URLs) is part of that one result — there are no add-on events billed on top.
• 🔓 Keyless, sign-free, proxy-free — no login, no API key, no signing keys, and no proxy needed. It just works from a default run.
• 🛡️ Real numbers, not deflated ones — TikTok serves flagged addresses a decoy version of a profile with wrong follower and view counts. This scraper is built so that never happens, so the stats you get are the stats TikTok shows a normal visitor. Profile counters (followerCount, heartCount, videoCount) come back exact, not rounded.
• 🧹 Clean, flat records — one tidy object per video, ready for sheets, dashboards or LLM pipelines. No nested junk to unwrap.
• 📬 Built-in Notion delivery — push every run into Notion via Apify connectors, zero extra code.
• 🚦 Honest errors — a clear errorCode row when something goes wrong, instead of silent emptiness. Diagnostic rows are never charged.
• ⏱️ Built-in cost guard — a run-time budget stops pathological queries from grinding on, so a bad keyword can't quietly run up a bill.

What you can scrape

You can mix all four in a single run — a list of profiles, a couple of hashtags and a search keyword all go in the same input and land in the same dataset.

Input

A real input

{
  "profiles": ["nasa", "nba"],
  "hashtags": ["spacex"],
  "searchQueries": ["rocket launch"],
  "startUrls": ["https://www.tiktok.com/@nasa/video/7300000000000000000"],
  "resultsPerPage": 50,
  "maxRunSeconds": 90
}

Input field names follow the conventions used across similar Actors, so switching over is usually a drop-in change — the same profiles / hashtags / searchQueries / startUrls shape you already have will run here without edits.

Output

One record per video:

{
  "ok": true,
  "type": "video",
  "id": "7300000000000000000",
  "webVideoUrl": "https://www.tiktok.com/@nasa/video/7300000000000000000",
  "text": "Liftoff! 🚀 #space",
  "createTime": "2024-06-12T14:00:00.000Z",
  "hashtags": ["space"],
  "region": "US",
  "playCount": 2400000,
  "diggCount": 318000,
  "commentCount": 4200,
  "shareCount": 9100,
  "collectCount": 15300,
  "downloadCount": 2100,
  "isAd": false,
  "video": {
    "duration": 23,
    "cover": "https://...jpg",
    "dynamicCover": "https://...webp",
    "playAddr": "https://...mp4",
    "downloadAddr": "https://...mp4"
  },
  "music": {
    "id": "6800000000000000000",
    "title": "original sound",
    "authorName": "NASA",
    "playUrl": "https://...mp3",
    "original": true,
    "duration": 23
  },
  "author": {
    "id": "6600000000000000000",
    "uniqueId": "nasa",
    "nickname": "NASA",
    "avatar": "https://...jpeg"
  }
}

And one record per profile, whenever you scrape by profile:

{
  "ok": true,
  "type": "profile",
  "id": "6600000000000000000",
  "uniqueId": "nasa",
  "secUid": "MS4wLjABAAAA...",
  "nickname": "NASA",
  "signature": "Explore the universe with us.",
  "verified": true,
  "avatar": "https://...jpeg",
  "followerCount": 9800000,
  "followingCount": 12,
  "heartCount": 210000000,
  "videoCount": 640,
  "diggCount": 31,
  "url": "https://www.tiktok.com/@nasa"
}

playAddr is the no-watermark MP4; downloadAddr is the watermarked one. Both are TikTok CDN links that expire after a while — download or rehost them if you need them long-term.

Filter the dataset by type (video or profile) if you want the two shapes in separate sheets, or use the CSV export, which flattens video.*, music.* and author.* into columns automatically.

Pricing

$0.25 per 1,000 results ($0.00025 each), with no run-start fee. Flat rate — no volume tiers, no plan gates — and you are charged only for results actually returned. The same $0.25 applies on every Apify plan, including the free one.

What is free:

• Diagnostic rows (ok: false with an errorCode) — you are never charged for a result you did not get.
• Runs that return nothing at all.
• Every field inside a result. Video stats, music/sound, author and profile fields are all part of the one result row; there is nothing billed as an extra event.

> Price change on 19 August 2026: the per-result price stays at $0.00025, and a $0.001 run-start fee is added to each run from that date. A 1,000-result run costs $0.25 today and $0.251 from that date — the extra $0.001 is charged once per run, however many results come back.

Because billing is per result, resultsPerPage is a hard cost ceiling: set it to 20 and a profile can never cost you more than $0.005.

Typical uses

• Influencer discovery — scrape a hashtag or keyword, sort by playCount or engagement rate (diggCount / playCount), then pull the author.uniqueId list back through the profiles input to get follower counts.
• Trend tracking — run the same hashtags on a schedule and diff playCount over time.
• Sound/audio research — group results by music.id to see which sounds are carrying a trend, and whether they are original.
• Competitor content audits — scrape a set of profiles weekly and chart posting cadence from createTime.
• Ad/creative sourcing — isAd flags promoted content, and video.cover gives you thumbnails for a contact sheet.

Rate limits and your own cookie

You do not need an account. Profiles, videos, hashtags and search are read from pages TikTok serves to any visitor, and that is the path every run takes first — it is the fastest one and it costs you nothing extra.

A handful of items are the exception: age-restricted videos, and pages TikTok occasionally decides to gate. For those the run falls back to a built-in session, which is shared across everyone using the Actor, so during a busy hour it can be rate-limited or temporarily unavailable. When that happens the run does not fail — it returns everything it did collect plus an uncharged diagnostic row for what it could not.

If you would rather not share that limit, paste your own TikTok cookie into sessionCookies:

1. Sign in to TikTok in Chrome. 2. F12 → Application → Cookies → https://www.tiktok.com. 3. Copy the sessionid value and pass it as "sessionid=<value>". Adding sid_tt and ttwid the same way ("sessionid=…; sid_tt=…; ttwid=…") makes it a little more durable.

Your cookie is used on its own — it is never mixed with the built-in session, so the limit you get is genuinely yours. It is sent only to TikTok's own pages, never to any other host, and its value is never written to the log. You can pass several cookies and the run will rotate between them. The run's status line reports sessionState, so you can see at a glance whether a run used the keyless path (keyless), a healthy session (ok), or ran out (exhausted).

Limitations — read these before you buy

Honest list. These are real, and knowing them up front saves you a wasted run.

• Public data only. Private accounts, follower/following lists, DMs and analytics behind a login are not returned by this Actor, whether or not you supply a cookie.
• Profile mode returns roughly the 10 newest videos per creator, plus the profile row. That is the whole of what TikTok publishes for a creator without a signed, logged-in request, so resultsPerPage above ~11 will not produce more rows for a single profile — put several handles in profiles instead. Hashtag and search modes page much deeper.
• No comment scraping in this Actor. Rows are videos and profiles only.
• Search and hashtag feeds can stall. TikTok sometimes keeps signalling "more pages" while returning near-empty ones. The Actor detects that, stops, and returns what it has — so a broad keyword can return fewer results than resultsPerPage even though results exist somewhere deeper.
• Throughput is paced. Requests are deliberately spaced to stay well inside upstream limits. Budget roughly 3-8 seconds per profile and one page of ~30 videos per second or two for hashtags, and size maxRunSeconds accordingly — split very large scrapes into several runs rather than one long one.
• Media URLs expire. playAddr, downloadAddr, cover and avatar are CDN links with a limited lifetime.
• Some fields can be null. region, music, secUid and the rarer counters (collectCount, downloadCount) depend on what TikTok exposes for that item; the Actor emits null rather than inventing a value.
• Counts are point-in-time. playCount and friends are whatever TikTok reported at the moment of the scrape.
• Video counters are at TikTok's display precision. For videos reached through profile mode or a /video/ URL, large numbers arrive rounded the way TikTok itself renders them (14,000,000 rather than 13,978,752) — that is the precision TikTok publishes on the page. Counters below about 100,000 are exact, and profile-level counters are exact at any size.

Errors and diagnostics

If something fails, the Actor still exits successfully and writes a diagnostic row instead of dying silently:

{ "ok": false, "errorCode": "NO_RESULTS", "error": "The request succeeded but returned no results for this input.", "mode": "search", "query": "..." }

Diagnostic rows carry ok: false — filter on that to separate them from real results in your pipeline. They are not charged.

Calling it from the API

curl -X POST "https://api.apify.com/v2/acts/Dami_Studio~tiktok-scraper/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"profiles":["nasa"],"resultsPerPage":20}'

That returns the dataset rows directly as JSON. For bigger jobs, start the run asynchronously and read the dataset when it finishes. The Actor also runs fine from Apify Schedules for recurring trend snapshots, and results can be pushed straight into Notion by supplying notionConnector.

FAQ

Do I need a TikTok account? No. Public profiles, videos, hashtags and search all work keyless, and that is what a default run uses. You can optionally paste your own cookie into sessionCookies to get a rate limit nobody else shares — see *Rate limits and your own cookie*.

Do I need a proxy? No. This scraper needs no proxy at all — that's a key advantage. Leave proxyConfiguration empty and it works.

Why do I sometimes see wrong follower counts on other tools? TikTok serves flagged addresses a decoy version of a profile with deflated stats. This Actor is built so that doesn't happen, so you get the same numbers a normal visitor sees.

It returned an errorCode row — why? That's the Actor telling you exactly what happened instead of failing silently. See the table above. You are not charged for it.

Can I get the video files? You get direct MP4 URLs (video.playAddr, no watermark). Downloading the files themselves is up to you — this Actor returns metadata and links, not binaries.

What does a 1,000-result run cost? $0.25 today. From 19 August 2026, $0.251 — the extra $0.001 is the run-start fee, charged once per run.