Synthesize and compare multiple LLM responses with OpenRouter council
Ready for 3.0Ready for 3.0. This template uses no nodes that n8n 3.0 removes. No removed nodes, versions or helpers found. Static check, last run 2026-10-19.
What this workflow does
This AI workflow has 20 nodes and runs when a chat message arrives. It uses only n8n's built-in nodes. It also calls openrouter.ai over HTTP.
How it runs
Trigger: When a chat message arrives.
- Initialize Variables (Edit Fields (Set)): sets 3 fields: userQuery, councilModels, chairmanModel
- Split by Model (Split Out): splits a list into separate items
- Stage 1: Query Model Answers (HTTP Request): calls openrouter.ai (POST)
- Stage 1: Extract Responses (Edit Fields (Set)): sets 2 fields: model, response
- Stage 1: Aggregate Responses (Aggregate): combines items into one
- Stage 2: Create Anonymized Responses (Edit Fields (Set)): sets 2 fields: labeledResponses, anonymizedText
- Stage 2: Build Ranking Prompt (Edit Fields (Set)): sets or maps fields
- Stage 2: Re-Add councilModels (Edit Fields (Set)): sets 2 fields: rankingPrompt, councilModels
- Stage 2: rankingPrompt per LLM (Split Out): splits a list into separate items
- Stage 2: Query Model Answer Rankings (HTTP Request): calls openrouter.ai (POST)
…plus 9 more steps. See the full node list below.
HTTP calls to: openrouter.ai
Before you upgrade to n8n 3.0
Nothing in this workflow is removed in n8n 3.0, so it should import and run after the upgrade.
Nodes used (7)
- Edit Fields (Set)
n8n-nodes-base.set - Split Out
n8n-nodes-base.splitOut - Chat Trigger
@n8n/n8n-nodes-langchain.chatTrigger - HTTP Request
n8n-nodes-base.httpRequest - Aggregate
n8n-nodes-base.aggregate - Send Email
n8n-nodes-base.emailSend - Code
n8n-nodes-base.code
From the author
This template adapts Andrej Karpathy’s LLM Council concept for use in n8n, creating a workflow that collects, evaluates, and synthesizes multiple large language model (LLM) responses to reduce individual model bias and improve answer quality. 🎯 The gist This…