Description
n8n-nodes-parallel
This is an n8n community node for Parallel Web Systems – the AI-powered web research platform. Transform your n8n workflows with intelligent data extraction, competitive analysis, lead enrichment, and automated research capabilities.
Try Parallel in the playground before integrating into your workflows.
n8n is a fair-code licensed workflow automation platform.
Installation
Install this community node package in n8n:
npm install n8n-nodes-parallel
Or follow the installation guide in the n8n community nodes documentation.
Prerequisites
1. Get API Access: Sign up at platform.parallel.ai and generate an API key
2. Add Credentials: In n8n, create new credentials of type “Parallel API” and enter your API key
3. Webhook Secret: In your Parallel n8n credential, include a webhook secret (created via Platform Settings) in order to be notified when an Async Web Enrichment Task is completed
Available Nodes
This package provides two powerful nodes for your n8n workflows:
🔍 Parallel Node – Multi-Purpose AI Research
The main Parallel node offers four distinct operations:
#### 1. Sync Web Enrichment – Task API
Execute tasks synchronously and get immediate results (up to 5 minutes):
- Lead Enrichment: Company profiles, contact information, financials
- Competitive Analysis: Product features, pricing, market positioning
- Content Research: Article summaries, fact-checking, data validation
- Market Intelligence: Industry trends, company updates, regulatory changes
- Complex multi-source research
- Deep competitive intelligence
- Comprehensive due diligence
- Large-scale data enrichment
- Natural language search objectives
- Traditional keyword queries
- Domain filtering for trusted sources
- Structured results with excerpts and citations
- Real-time web-informed responses
- Fact-checking and verification
- Current events and news queries
- Research-backed conversations
- Real-time notifications: Instant updates when async tasks finish
- Automatic result fetching: Retrieves full task results automatically
- Webhook security: Built-in signature validation for secure webhooks
- Flexible filtering: Option to trigger only on successful completions
- Text: Single text output field with evidence and citations
- JSON: Custom structured schemas for precise data extraction
- Auto: AI-optimized JSON output with nested citations (Pro+ processors only)
- Automatically handles long-running research tasks with intelligent retry logic
- Webhook triggers for immediate notification when async tasks complete
- No manual polling required
Available Processors: lite, base, core
#### 2. Async Web Enrichment – Task API
Start long-running research tasks asynchronously (up to 30 minutes):
Available Processors: lite, base, core, pro, ultra, ultra2x, ultra4x, ultra8x
Learn about each Parallel Task API Processor here.
#### 3. Web Search – Search API
Intelligent web search with AI-powered processing:
Available Processors: base, pro
Learn about each Parallel Search API Processor here.
#### 4. Web Chat
AI-powered chat completions with web access (< 5 seconds):
Read the Parallel Chat API documentation here.
🔔 Parallel Task Run Completion Trigger
Automatically trigger workflows when Parallel tasks complete, best paired with Async Web Enrichment:
Perfect for long-running research workflows where you want to process results as soon as they’re ready.
Key Features
Flexible Output Schemas
Research Quality & Transparency
Every result includes confidence scores and evidence with source citations, so you can validate and trust your automated research.
Source Control
Use source policies to include trusted domains (like Wikipedia, Reuters) or exclude unreliable ones.
Smart Polling & Async Support
Quick Start Examples
Example 1: Company Research Workflow
1. HTTP Request node → Get company domain from CRM
2. Parallel node (Sync Web Enrichment) → Enrich company data
3. Set node → Format results
4. HTTP Request node → Update CRM with enriched data
Example 2: Async Research with Webhook
1. Parallel node (Async Web Enrichment) → Start deep research task
2. Parallel Task Run Completion Trigger → Wait for completion
3. Code node → Process comprehensive results
4. Email node → Send research report
Example 3: Web Search Pipeline
1. Manual Trigger → Input search query
2. Parallel node (Web Search) → Search with AI processing
3. Code node → Extract top results
4. HTTP Request → Post to knowledge base
Getting Started
1. Install: npm install n8n-nodes-parallel
2. Get API Access: Sign up at platform.parallel.ai and generate an API key
3. Try the Playground: Test your research tasks at platform.parallel.ai/play
4. Add Credentials: Configure your API key in n8n’s credential manager
5. Build Workflows: Start with simple enrichment tasks and scale to complex research pipelines
Processor Selection Guide
Choose the right processor based on your use case complexity and timing needs:
| Processor | Latency | Max Fields | Cost | Best For |
|———–|———|————|——|———-|
| Lite | 5s-60s | 2 | $5/1000 | Basic metadata, quick lookups |
| Base | 15s-100s | 5 | $10/1000 | Standard enrichment, reliable data |
| Core | 1-5min | 10 | $25/1000 | Cross-referenced research |
| Pro | 3-9min | 20 | $100/1000 | Exploratory research, analysis |
| Ultra | 5-25min | 20 | $300/1000 | Advanced multi-source research |
| Ultra 2x | 5-25min | 25 | $600/1000 | Difficult research tasks |
| Ultra 4x | 8-30min | 25 | $1200/1000 | Very difficult research |
| Ultra 8x | 8-30min | 25 | $2400/1000 | Most complex research |
> Note: Sync operations support up to Core level. Async operations support all processor levels.
n8n Workflow Use Cases
Sales & CRM Automation
Content & Marketing
Operations & Monitoring
Customer Support Enhancement
Advanced Configuration
Webhook Setup for Async Tasks
When using async enrichment with the trigger node:
1. Add a Parallel Task Run Completion Trigger to your workflow
2. Copy the webhook URL from the trigger node
3. In the Parallel node (Async Web Enrichment), paste the URL in the “Webhook URL” field
4. Enable “Validate Webhook Signatures” for security (recommended)
JSON Schema Examples
For structured data extraction, use JSON schemas in the output configuration:
{
"type": "object",
"properties": {
"company_name": {
"type": "string",
"description": "Official company name from recent filings or website."
},
"ceo_name": {
"type": "string",
"description": "Current CEO full name from company website or recent news."
},
"employee_count": {
"type": "string",
"description": "Number of employees as range (e.g., '500-1000') or exact number."
}
},
"required": ["companyname", "ceoname", "employee_count"],
"additionalProperties": false
}
Source Policies
Control which websites are used for research:
wikipedia.org, reuters.com, bloomberg.comreddit.com, quora.com
