Verifai

What is Verifai?

VerifAI’s MultiLLM is an open-source Python framework that runs multiple language models (LLMs) in parallel, compares their outputs, and ranks the results to identify the most accurate answers. It is built primarily for comparing code generated by popular LLMs such as GPT3, GPT5, and Google-Bard. You can add new LLMs and customize how outputs are ranked. Cross-checking results across models reduces the risk of relying on incorrect information from a single LLM. The framework is available on GitHub.

What is Verifai used for?

• Comparing code written by popular LLMs like GPT3, GPT5, and Google-Bard
• Ranking responses from multiple LLMs for code generation or answers to specific questions
• Adding support for new LLMs and customizing how results are ranked
• Cross-checking outputs from several models to reduce the risk of trusting incorrect information

Who is Verifai for?

Verifai is aimed at software developers, data scientists, researchers, academics, and students.

Getting started with Verifai

Install the MultiLLM framework from GitHub along with its dependencies. Configure it to run your chosen LLMs in parallel and adjust the ranking functions to evaluate outputs according to your needs. Submit a task — code generation or a general question — and MultiLLM ranks the outputs to surface the most accurate result. The Verifai blog article on MultiLLM covers more detail, and the open-source repository tracks new features and improvements.