Fast Google Patents Scraper — Prior Art & Claims
Pricing
Pay per usage
Fast Google Patents Scraper — Prior Art & Claims
Ultra-fast Google Patents scraper without browser overhead. Extract global patents, publication numbers, priority dates, abstracts, claims, inventors, corporate assignees, and direct PDF download links across USPTO, EPO, WIPO, and 100+ patent offices on 256MB RAM.
Pricing
Pay per usage
Rating
0.0
(0)
Developer
Hector
Maintained by CommunityActor stats
1
Bookmarked
2
Total users
1
Monthly active users
3 days ago
Last modified
Categories
Share
⚡ Quick Test (0-Setup): Click the Input tab above or Try for free — all fields are pre-filled with a real-world example. Simply click Start to see live verified data in seconds! 💡 Ultra-lightweight: Runs on 256 MB RAM — costs < $0.001 per run. 100% Free test included with your Apify monthly credits!
Zero-proxy, sub-second Google Patents prior art, claims & PDF scraper on 256MB RAM
⚡ Live Run Demo (Sub-second Execution)
[1;32m[00:00.00][0m 🚀 Initializing Micro-Container (256MB RAM)...[1;34m[00:00.24][0m 🛡️ Anti-bot handshake passed: Native protocol bypass (Zero-Proxy Required)[1;36m[00:00.68][0m ⚡ Scraping Google Patents Global IP Intelligence... 50 records extracted & sanitized[1;32m[00:01.35][0m ✅ STATUS: SUCCEEDED in 1.35s — Compute cost: $0.00078[1;35m[00:01.36][0m 📦 Export ready: [JSON] [CSV] [Excel] [Webhook Delivered]
🛡️ 100% Zero-Block & Zero-Proxy Guarantee
- 🟢 Status:
Operational & Actively Maintained(Verified daily for October 2026 anti-bot rules).- ⚡ Smart Protocol Bypass: Queries official internal JSON & RPC protocols directly with genuine client headers. Zero heavy Chrome overhead, zero Cloudflare/Datadome 403 blocks.
- 🚫 Zero Expensive Proxies Needed: Runs effortlessly on direct connections or standard Apify proxies. Saves you $15/GB compared to bloated headless browser scrapers.
📊 Data Sneak Peek: Clean Spreadsheet View (Excel & Google Sheets Compatible)
| publicationNumber | title | assignee | inventor | filingDate | publicationDate | pdfUrl |
|---|---|---|---|---|---|---|
US11650869B2 | Quantum computing service with local edge devices | Amazon Technologies, Inc. | Jeffrey Paul Heckey | 2019-11-27 | 2023-05-16 | https://patentimages.storage.googleapis.com/.../US11650869.pdf |
🎯 3 One-Click Real-World Scenarios
Choose your use case and launch it in 1 click:
- 🎯 Scenario 1: Freedom-to-Operate & Prior Art Search — Audit existing international patents across USPTO, EPO, and WIPO before filing new intellectual property.
- 🎯 Scenario 2: Competitor R&D Technology Landscaping — Track secret corporate patent filings from Apple, Tesla, or Nvidia months before products hit the public market.
- 🎯 Scenario 3: Automated Patent Legal Briefs (RAG) — Pipe claims and abstracts into LLMs to automatically generate technical patent infringement matrices.
👉 Open the Input Form to Launch Your Free Test
💰 Transparent Cost Breakdown: What Does 1 Run Cost?
Unlike bloated scrapers that burn through monthly credits in 10 searches, this Actor is engineered for ultra-lightweight 256MB micro-containers:
| Metric | Heavy Competitor Scrapers | Fast Scraper (This Actor) |
|---|---|---|
| RAM Allocated | 2048 MB – 4096 MB | 256 MB |
| Average Run Time | 25 – 60 seconds | Sub-second to 2.5 seconds |
| Compute Cost per Run | ~$0.05 per batch | ~$0.0008 (Less than 1/10th of a cent) |
| Apify Free Tier ($5/mo) | ~100 runs before credits run out | 🔥 Up to 6,000+ runs 100% FREE every month |
🎁 Every Apify account includes $5.00 in free compute credits every single month. You can run and test this scraper completely free without entering a credit card or upgrading.
💻 Developer API Integration (Python & Node.js)
Call this Actor programmatically from your backend, ETL pipelines, or LLM agents in 3 lines of code:
🐍 Python SDK
from apify_client import ApifyClient# Initialize the client with your Apify API tokenclient = ApifyClient("YOUR_APIFY_TOKEN")# Run the Actor and wait for completionrun = client.actor("rbbscrap/fast-google-patents-scraper").call(run_input={"searchQuery": "solid state battery lithium ceramic","countryCode": "US","status": "GRANT","maxItems": 50})# Fetch dataset items directly as Python dictionariesitems = client.dataset(run["defaultDatasetId"]).list_items().itemsprint(f"Extracted {len(items)} items successfully!")
🟡 Node.js / TypeScript SDK
import { ApifyClient } from 'apify-client';const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });const run = await client.actor('rbbscrap/fast-google-patents-scraper').call({searchQuery: "solid state battery lithium ceramic",countryCode: "US",status: "GRANT",maxItems: 50});const { items } = await client.dataset(run.defaultDatasetId).listItems();console.log(`Extracted ${items.length} items successfully!`);
⚡ Integrations & Export Formats
Export your datasets in any format in 1 click:
- Formats: JSON, CSV, Microsoft Excel (.xlsx), XML, RSS, HTML Table.
- No-Code Automations: Connect directly to Make.com, Zapier, n8n, and Google Sheets.
- Webhooks: Configure instant webhooks to ping your backend server upon completion.