Spark
Sign In
MCP Connector

Query Multiple AI Models for Diverse Insights

Multi-Model Advisor queries several local Ollama models on one question and combines their answers into an expert council.

Connect
Works with ollama
YuChenSSR
YuChenSSR
95
Spark score
out of 100
Status Verified
Updated 5 months ago
Version 1.0.0
Models
universal
Add to Favorites

Why it matters

Leverage an 'expert council' of multiple Ollama models to generate diverse perspectives on a single query. This asset combines AI responses, allowing for richer analysis and decision-making.

Outcomes

What it gets done

01

Query multiple Ollama models simultaneously with a single prompt.

02

Assign custom roles and system prompts to individual AI models.

03

View and manage all available Ollama models on your system.

04

Integrate seamlessly with Claude for Desktop for enhanced workflows.

Install

Add it to your toolbox

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/vb-multi-model-advisor | bash

Capabilities

Tools your agent gets

list-available-models

Shows all Ollama models available on your system

query-models

Queries multiple Ollama models with a single question and combines their responses

Overview

Multi-Model Advisor MCP Server

Multi-Model Advisor queries multiple local Ollama models with one question and combines their answers, with per-model persona customization. Use it when you want to compare how multiple local Ollama models answer the same question, each optionally with its own persona.

What it does

Multi-Model Advisor is an MCP server that queries multiple local Ollama models on a single question and combines their responses, giving you an "expert council" of diverse AI perspectives instead of relying on just one model's answer.

When to use - and when NOT to

Use it when you want to compare how different local Ollama models respond to the same question, optionally assigning each one a different role or persona to draw out varied angles, for example asking Gemma, Llama, and DeepSeek to each weigh in on the most important skills for career success. It requires Ollama installed and running locally with the models you want to query already pulled, so it is not usable without a local Ollama setup and pulled model weights. It is a comparison and diversity tool, not a routing layer that picks the single "best" model automatically; you see all the responses and judge them yourself.

Capabilities

list-available-models shows all Ollama models currently on your system. query-models sends a single question to multiple models at once and returns their combined responses, with each model's system prompt customizable independently via environment variables like GEMMA_SYSTEM_PROMPT, LLAMA_SYSTEM_PROMPT, and DEEPSEEK_SYSTEM_PROMPT, so you can give each model a distinct role or persona for the same question.

How to install

Via Smithery, for Claude Desktop:

npx -y @smithery/cli install @YuChenSSR/multi-ai-advisor-mcp --client claude

Or from source, which also pulls the default models used in examples:

git clone https://github.com/YuChenSSR/multi-ai-advisor-mcp.git
cd multi-ai-advisor-mcp
npm install
npm run build
ollama pull gemma3:1b
ollama pull llama3.2:1b
ollama pull deepseek-r1:1.5b

Configure Claude Desktop to run the built index.js via node. Optional environment variables let you set the server name and version, enable debug mode, point at a custom OLLAMA_API_URL, and set DEFAULT_MODELS as a comma-separated list. Requires Node.js 16+ and a running local Ollama installation.

Who it's for

Anyone who wants a quick, local comparison of how multiple small open models answer the same question, without paying for or calling out to cloud model APIs.

Source README

An MCP server that queries multiple Ollama models and combines their responses, providing diverse AI perspectives on a single question through an "expert council" approach.

Installation

Smithery

npx -y @smithery/cli install @YuChenSSR/multi-ai-advisor-mcp --client claude

From Source

git clone https://github.com/YuChenSSR/multi-ai-advisor-mcp.git
cd multi-ai-advisor-mcp
npm install
npm run build
ollama pull gemma3:1b
ollama pull llama3.2:1b
ollama pull deepseek-r1:1.5b

Configuration

Claude Desktop

{
  "mcpServers": {
    "multi-model-advisor": {
      "command": "node",
      "args": ["/absolute/path/to/multi-ai-advisor-mcp/build/index.js"]
    }
  }
}

Available Tools

Tool Description
list-available-models Shows all Ollama models on your system
query-models Queries multiple models with a single question

Features

  • Query multiple Ollama models with a single question
  • Assign different roles/personas to each model
  • View all available Ollama models on your system
  • Customize system prompts for each model
  • Configuration via environment variables
  • Seamless integration with Claude for Desktop

Environment Variables

Optional

  • SERVER_NAME - Server name
  • SERVER_VERSION - Server version
  • DEBUG - Enable debug mode
  • OLLAMA_API_URL - URL for Ollama API
  • DEFAULT_MODELS - Comma-separated list of default models
  • GEMMA_SYSTEM_PROMPT - System prompt for Gemma model
  • LLAMA_SYSTEM_PROMPT - System prompt for Llama model
  • DEEPSEEK_SYSTEM_PROMPT - System prompt for DeepSeek model

Usage Examples

Show me which Ollama models are available on my system

what are the most important skills for success in today's job market, you can use gemma3:1b, llama3.2:1b, deepseek-r1:1.5b to help you

Notes

Requires Ollama to be installed and running locally. Each model can be assigned different personas/roles to obtain diverse perspectives. Supports Node.js 16.x or higher.

FAQ

Common questions

Trust

How it checks out

Official By maintainer
Downloads 2

Discussion

Questions & comments · 0

Sign In Sign in to leave a comment.