Which models are compatible with the Multi-Model skill?

The skill supports any OpenAI-compatible API, including cloud providers like OpenAI and local instances running via Ollama.

Is my sensitive information hidden from the models?

Yes, the skill is designed to treat model identities as numeric IDs and explicitly prevents the printing or sharing of API keys found in your environment files.

How does the 'battle' mechanism improve results?

One model generates an initial response while others critique it. The process rotates roles over several iterations, keeping only the highest-scoring output from each round.

Can I use this skill with local models to save costs?

Yes, by configuring the skill to point to a local Ollama endpoint, you can run multi-model refinements entirely on your own hardware.

What is recursive reasoning in this context?

It refers to the iterative process where each model builds upon and improves the best result from the previous turn, leading to a more polished final output.

Multi-Model Battle & Refinement

Name: Multi-Model Battle & Refinement
Author: lollipopkit

bylollipopkit

•

Data Science & ML

Orchestrates iterative refinement cycles across multiple LLMs to produce high-quality, peer-reviewed outputs through competitive judging.

The Multi-Model skill enables Claude to facilitate an automated 'model arena' directly within your development environment. By rotating through a selection of OpenAI-compatible or Ollama models, the skill manages a recursive reasoning loop where different LLMs generate candidate answers and peer-critique each other's work. This process ensures that the final output is the result of multiple layers of cross-model evaluation, making it an essential tool for complex tasks that require high accuracy, rigorous logic, or creative diversity.

Key Features

01Automated cross-model judging and scoring to select superior responses

02Configurable iteration depth and judge limits for optimized performance

035 GitHub stars

04Iterative model rotation for progressive output refinement

05Support for local Ollama instances and remote OpenAI-compatible providers

06Privacy-centric architecture that masks model identities and protects API secrets

Use Cases

01Benchmarking specific model strengths and weaknesses on domain-specific tasks

02Comparing code implementations across different LLMs to find the most efficient solution

03Refining complex documentation or prompts through recursive model peer-review

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add lollipopkit/cc-plugins multi-model

For use in Claude.ai and ChatGPT

Download Skill