Overview

cx-cmu/AutoGEO_mini_Qwen1.7B_GEOBench is a 2 billion parameter model built on the Qwen1.7B architecture, developed by cx-cmu. It functions as a GEO (Generative Engine Optimization) model, designed to optimize web documents for better integration and visibility within large language model-based generative search engines. The core purpose is to rewrite documents to align with the preferences of generative engines (e.g., GPT, Gemini, Claude), thereby increasing their likelihood of being included in generated responses, all while maintaining the original factual accuracy and meaning.

Key Capabilities

• Document Rewriting: Rewrites web documents to make them more appealing to generative AI engines.
• Visibility Enhancement: Aims to increase the "visibility and coverage" of web content in AI-generated responses.
• Meaning Preservation: Ensures that the original meaning and factual content of the document are retained during the rewriting process.
• Targeted Optimization: Specifically trained and optimized for use with gemini-2.5-flash-lite on the GEO-Bench dataset.

Important Considerations

• This model is part of the broader AutoGEO framework, which provides the tools and methodology for its application.
• For use with other generative engines or datasets, post-training of Qwen/Qwen3-1.7B using the AutoGEO code is recommended.

Good for

• Developers and content creators looking to optimize web documents for generative search engine visibility.
• Research into how generative AI engines process and utilize web content.
• Integration into systems that aim to improve the discoverability of information through LLM-powered search.