Parallel icon

Parallel

Web Actions - Web Enrichment and Web Search powered by AI
Join our community

Actions2

Overview

This node, named "Parallel," provides AI-powered web actions focused on two main operations: Web Enrichment and Web Search. The Web Enrichment operation allows users to execute tasks that involve AI-driven web research and data extraction based on a given input query or JSON object. It can return results either as plain text or structured JSON according to a user-defined schema. This is useful for scenarios such as extracting specific factual information from the web, summarizing data points, or generating structured insights from online content.

For example, you could ask the node: "What was the GDP of France in 2023?" and receive a formatted textual answer or a JSON object containing the relevant data. The node supports customization of the output format and filtering of source domains to tailor the enrichment process.

Properties

Name Meaning
Input The input query or JSON object describing the task to perform. For example, a question like "What was the GDP of France in 2023?"
Output Schema Type Defines the desired output format:
- Text: A plain text description of the result.
- JSON: A structured JSON object following a user-provided schema.
Output Description (Shown if Output Schema Type is Text) A textual description guiding the expected output format, e.g., "GDP in USD for the year, formatted like "$3.1 trillion (2023)"".
JSON Schema (Shown if Output Schema Type is JSON) A JSON schema defining the structure and validation rules for the expected output JSON object.
Processor Selects the processing engine:
- Lite: Basic metadata, fallback option with low latency.
- Base: Standard reliable enrichments with higher cost per run.
Additional Fields Optional extra parameters:
- Include Domains: Comma-separated list of domains to restrict search results to.
- Exclude Domains: Comma-separated list of domains to exclude.
- Input Schema: Optional description of expected input.
- Metadata: Custom key-value pairs to store with the run.

Output

The node outputs an array of JSON objects, each representing the result of the enrichment task for one input item.

  • The json field contains the response returned by the external API:

    • If the output schema type is Text, this will be a plain text string describing the result.
    • If the output schema type is JSON, this will be a JSON object conforming to the user-defined schema.

  • No binary data output is produced by this node.

Dependencies

  • Requires an API key credential for authentication with the external Parallel AI service.
  • The node makes HTTP requests to https://api.parallel.ai endpoints.
  • Proper configuration of the API credential in n8n is necessary for successful execution.

Troubleshooting

  • Invalid JSON in Output Schema: If the JSON schema provided is not valid JSON, the node will throw an error indicating invalid JSON in the output schema. Ensure the JSON schema is correctly formatted.
  • Task Execution Timeout: The node attempts to retrieve the task result multiple times (up to 15 attempts). If the task does not complete within this time frame (~4 minutes per attempt), it throws a timeout error. Consider simplifying the input or using the "Lite" processor for faster responses.
  • API Request Failures: Errors during API calls will be surfaced as node operation errors. Check your API key validity, network connectivity, and ensure the Parallel AI service is operational.
  • Domain Filtering Issues: Incorrectly formatted domain lists in "Include Domains" or "Exclude Domains" may lead to unexpected results or no results. Use comma-separated domain names without spaces.

Links and References

Discussion