Loop Mode allows nodes to process lists of inputs automatically, like a “for loop” in programming. Instead of processing one item at a time, nodes in Loop Mode can handle multiple items efficiently.

​
Basic Concept

When you enable Loop Mode on a node:

  • The node processes each item in the list individually
  • The output is a list, with each item representing the corresponding input

​
Example

Summarizer Node:
Normal Mode: One email → One summary
Loop Mode: [email1, email2, email3] → [summary1, summary2, summary3]

​
Concurrent Processing

Loop Mode can process multiple items simultaneously based on your plan:

  • Free Plan: 2 concurrent items
  • Starter Plan: 5 concurrent items
  • Pro Plan: 15 concurrent items

For example, with 15 concurrent steps:

Input: 100 websites to scrape
Processing: 15 websites processed simultaneously
Result: Faster completion through parallel processing

​
Best Practices

​
1. Error Protection

Always wrap nodes in Loop Mode with Error Shield to:

  • Prevent entire workflow failure if one item fails
  • Collect successfully processed items
  • Track failed items for troubleshooting

Example:

Web Scraper (Loop Mode) inside Error Shield
Input: [url1, url2, url3]
- Success: [content1, content3]
- Errors: [url2]

​
2. Handling List Size Mismatches

The problem occurs when:

  1. Your node has multiple inputs
  2. You’re connecting values dynamically (from other nodes)
  3. The connected lists have different sizes

Example Problem:

Ask AI node with:
- Prompt from previous node: [Single Prompt]
- Articles from sheet: [Article1, Article2, Article3]
❌ Error! Lists don't match (1 ≠ 3)

Solution: Use the Duplicate node to match list sizes:

  1. Connect your smaller list to a Duplicate node
  2. Connect your larger list size to the list size to match input of the duplicate node
  3. Connect the result to your node
[Single Prompt] → Duplicate (count: 3) → Ask AI
[Article1, Article2, Article3] → Ask AI
Output: AI response for all three Articles

​
Additional Tips

  1. Use subflows to avoid “list of lists” issues
  2. Consider concurrent processing limits for your plan for large loop operations

​
Example Flow

In this example, the Summarizer node processes multiple emails in Loop Mode, creating a summary for each email automatically.

SubflowsCredits