Facebook Reels & Video Scraper — MP4 Download, Transcript, No Login

Scrape and download public Facebook Reels and Videos without login or cookies. Give the scraper a Reel/Video URL — or just a keyword — and get back the direct MP4 download link, thumbnail, caption, hashtags, view count, reactions, comments, shares, creator details and a clean transcript / subtitles. Export to JSON, CSV or Excel.

Built for predictable cost and reliability: bounded result budgets, successful-video events and clean run-and-exit. Platform usage and optional external ASR remain separate costs.

---

✅ What this Facebook video scraper does

• 📥 Download Facebook Reels & Videos (MP4) — available HD/SD links for supported public videos
• 📝 Transcript & subtitles — native captions parsed to text + timed cues (free), or bring your own speech-to-text key
• 🔎 Keyword & hashtag search — discover public Reels/Videos via search engines, no URL list needed
• 📊 Full engagement data — views, reactions, comments, shares + a 0–100 engagement score
• 👤 Creator info — name, profile URL, ID, followers, rating
• 🏷️ Smart metadata — hashtags, @mentions, auto category, language, music/audio, resolution & orientation
• 📅 Date filter, geo-targeting, real-time webhook, retries — control that most scrapers skip

No login, no cookies, no API key. Only public content.

📦 Output (per video)

Field	Description
videoId, videoUrl, videoType	ID, canonical URL, reel/video/live
title, caption, hashtags, mentions, category, language	Content & metadata
videoMp4Url (+ HD/SD), thumbnailUrl, durationSeconds	Media & download links
width, height, aspectRatio, orientation	Video dimensions
viewCount, reactionCount, commentCount, shareCount, engagementScore	Engagement; score is null when views or interaction counts are unavailable
creatorName, creatorUrl, creatorId, creatorFollowers, creatorRating	Creator
publishedAt, musicTitle, musicArtist	Timing & audio
transcript, transcriptCues, transcriptSummary, hasTranscript	Transcript

⚙️ Input

Field	What it does
workflow	Auto-detect, specific video URLs, keyword search, profiles, or combined sources
startUrls	Reel/Video URLs — /reel/<id>, /watch/?v=<id>, /<page>/videos/<id>, fb.watch/…, bare IDs
searchQueries	Keywords / #hashtags to discover public Reels & Videos
profileUrls	Facebook Profiles or Pages whose public reels should be discovered
maxResults, maxCostUsd	Result ceiling and estimated video-event budget; not an all-in invoice cap
outputPreset	compact for MCP, engagement for social metrics, or legacy-compatible full
sinceDays, keywordFilter	Keep only recent videos / captions matching a keyword
downloadMp4, includeTranscript	Toggle MP4 links and native transcript
transcribeWithAsr + asrApiKey	Optional: transcribe audio with your own OpenAI-compatible speech-to-text key
proxyCountry	Geo-target the proxy (US, GB, DE, …)
maxConcurrency, timeout, maxRetries	Throughput & reliability
notifyWebhookUrl	POST each result to Zapier / Make / n8n / Slack in real time

Example input

{
  "workflow": "combined",
  "searchQueries": ["#fitness"],
  "startUrls": ["https://www.facebook.com/reel/2166091230582141/"],
  "maxResults": 10,
  "outputPreset": "engagement",
  "sinceDays": 30,
  "downloadMp4": true,
  "includeTranscript": true,
  "proxyCountry": "US"
}

MCP and AI-agent quick start

Use an explicit workflow, a small maxResults, and a compact preset so the tool call is predictable and the response stays small:

{
  "workflow": "videoUrls",
  "startUrls": ["https://www.facebook.com/reel/2166091230582141/"],
  "maxResults": 3,
  "outputPreset": "compact"
}

Read SUMMARY after every run. It contains status, workflow, output preset, charged row count, candidate coverage and next steps. Read ERRORS for invalid input and skipped or failed candidates. outputPreset: "full" remains the default for backward compatibility; the current schema and omitted-input runtime default for maxResults are 1,000. Set a smaller explicit ceiling for a first run; zero disables that result ceiling.

Supported workflow capabilities

Capability	Coverage to check
URL or ID with video metadata	Public availability and canonical identity
MP4 links	Available media formats; links can expire
Native captions or optional ASR	Source captions, or the user's own external service
Keyword discovery	Query relevance and bounded candidate coverage
Recency filtering	Available publication timestamps
Category, music and creator metadata	Optional or derived fields; coverage varies
Proxy country	Request region, not verified viewer geography
Webhook delivery	User-configured endpoint and its delivery status
Video-event budget	Estimated event budget; resources are separate

💡 Use cases

• Download Facebook Reels in bulk for re-use, archives or analysis
• Video transcript extraction for SEO, captions, repurposing to blog/LinkedIn
• Competitor & creator research — track engagement, hooks and posting cadence
• Trend & hashtag monitoring — find top-performing Reels in any niche
• Ad creative & influencer scouting — filter by engagement, followers, recency
• Feeding n8n / Make / Zapier automations and AI agents (MCP)

💸 Pricing

Pay per result. Video events are charged for successfully scraped and saved videos; filtered, duplicate, wrong-profile or failed candidates do not receive a video event. Actor-start and Apify platform usage charges may still apply. maxCostUsd is an estimated video-event budget, not an exact total invoice cap.

❓ FAQ

Do I need to log in or provide cookies? No. The scraper only reads public Reels and Videos.

Can I download the MP4? Yes — videoMp4Url gives a direct HD/SD link. Facebook links are time-limited, so download promptly.

Paste an actual Facebook URL, not a placeholder ID. Links copied as Markdown ([text](https://...)) are normalized automatically, and duplicate Reel/Watch forms of the same video are fetched once.

Does it get the transcript? Yes, when the video has captions (parsed to text). For videos without captions, enable BYO speech-to-text with your own API key.

Can I search instead of providing URLs? Yes — use searchQueries with keywords or #hashtags.

Search candidates must also match your query in available video title, caption, creator name or hashtags. Unrelated or unverifiable fallback results are skipped without a video event; this can reduce the count but prevents charging for irrelevant videos.

How can I get more results? Increase maxResults and, for profiles, maxReelsPerProfile. searchDepth explores more public search pages; candidateMultiplier reserves candidates to replace unavailable or rejected videos. Defaults balance coverage and requests. Higher settings cost more and cannot guarantee the requested count or a complete profile archive. Profile matches use owner URL/ID when exposed, or exact owner name with an ID; ambiguous matches are skipped.

How do I control costs? Use maxResults, the estimated maxCostUsd budget and Apify run charge/time limits. Start and platform usage charges are separate from video events.

What about private videos? Only public content is returned. Private/removed videos are reported with a clear status and are not billed.

The Dataset contains only delivered videos, so one video equals one Dataset row. Run totals are available separately in the SUMMARY record; failed candidate diagnostics are stored in ERRORS and do not receive a video event.

---

Not affiliated with or endorsed by Facebook/Meta. Use responsibly and in line with applicable laws and platform terms; scrape only public data.

Ready-to-import n8n workflow

Monitor public videos and hand normalized results to the next step with the maintained Facebook Reels n8n connector. Download the quickstart workflow, import it into n8n, add your Apify API credential, and provide the public URLs you want to process.

Plan a Facebook Reels research workflow

A Facebook Reels scraper can support creator research, content inventories and video analysis when the selected public videos expose enough information. Start with a known public video before running broad keyword discovery. Keep canonical video IDs with your export so that an expiring media link is not your only reference to an observation.

Research goal	Useful settings	Review before using the result
Inventory a public video	Explicit URL and compact output	Canonical ID, caption and media coverage
Compare creator content	Profile workflow and a small result limit	Owner match and incomplete discovery coverage
Explore a topic	Search query and engagement output	Query relevance and available count fields
Process captions	Native transcript enabled	hasTranscript, language and missing caption tracks

Inspect one public video

{
  "workflow": "videoUrls",
  "startUrls": [
    "https://www.facebook.com/reel/2166091230582141/"
  ],
  "maxResults": 1,
  "outputPreset": "compact",
  "downloadMp4": false,
  "includeTranscript": false
}

This is a small metadata request. Disabling MP4 delivery does not request a downloaded video file or guarantee that a public video remains accessible.

Build a small creator sample

{
  "workflow": "profiles",
  "profileUrls": [
    "https://www.facebook.com/NASA"
  ],
  "maxReelsPerProfile": 3,
  "maxResults": 2,
  "outputPreset": "engagement",
  "downloadMp4": false,
  "includeTranscript": false
}

Profile discovery is a sample of available candidates. It is not an exhaustive archive, and an owner mismatch is skipped. An empty result is not evidence that the page has never published a Reel.

Research a specific public topic

{
  "workflow": "search",
  "searchQueries": [
    "NASA mission"
  ],
  "maxResults": 2,
  "searchDepth": 1,
  "candidateMultiplier": 2,
  "outputPreset": "engagement",
  "downloadMp4": false
}

Candidates are checked against the query in available content or creator fields. A limit is a ceiling, not a guaranteed quantity or a popularity ranking. Search-engine coverage can lag the original video.

Request native captions without an external ASR key

{
  "workflow": "videoUrls",
  "startUrls": [
    "https://www.facebook.com/reel/2166091230582141/"
  ],
  "maxResults": 1,
  "includeTranscript": true,
  "transcribeWithAsr": false,
  "downloadMp4": false,
  "outputPreset": "full"
}

When native captions are unavailable, hasTranscript can be false. Empty transcript text is not a silent video. Optional external ASR is a separate service requiring the user’s own key and potentially its own charges.

Budget, export and interpretation

The current video-event rates per 1,000 delivered videos are $5 Free, $4 Bronze, $3.50 Silver, $3 Gold/Platinum and $2.50 Diamond. A startup event costs $0.00005 per GB with a minimum of one event; platform usage is additional. For example, two videos at the Free result rate cost $0.01 in video events, before startup, platform resources and any external speech service. Use the live Pricing tab for the account's applicable rate.

maxCostUsd estimates the video-event budget using an internal reference rate. It does not cap every component of the bill, particularly across plan tiers. Use an Apify run spending limit and a timeout alongside small result limits. Larger search depth and candidate reserves may consume requests without adding any delivered rows. Increasing concurrency cannot make blocked content public.

Inspect SUMMARY and ERRORS before handing the Dataset to n8n or an AI agent. Keep videoId/videoUrl, collected data and coverage together; deduplicate future exports by video identity. JSON preserves timed cues and nested metadata more faithfully than a single CSV cell. A returned MP4 URL can expire and is not a permanent file in your storage. Scores compare available observations, not verified reach, unique viewers, advertising conversion or influencer quality.

Does a Reel always have HD media, a transcript and all engagement counts? No. Those fields depend on public source coverage. Missing counts are unknown; missing media or captions must not be replaced by fabricated values.

Does a profile or hashtag search return every matching video? No. Discovery has bounded candidate coverage, filtering and source availability constraints. Keep the requested workflow and the returned candidate status with the export.

n8n and developer integrations

Use this Actor in your automation or application with the existing community integrations below. An Apify account and API token are required; Actor and platform charges follow the current pricing.

• n8n community node: n8n-nodes-apivault-facebook-reels. Install package name n8n-nodes-apivault-facebook-reels in a self-hosted n8n instance that permits community packages.
• n8n source and usage: GitHub.
• Python client and examples: GitHub.
• n8n source and usage: GitHub.

These are community packages. Publication on npm does not establish n8n verification or availability in n8n Cloud. Review the linked package README for supported inputs and setup.