Patronus AI Launches Industry-First Multimodal LLM-as-a-Judge for Image Evaluation

Patronus AI today announced the launch of the industry’s first Multimodal LLM-as-a-Judge (MLLM-as-a-Judge), a groundbreaking evaluation capability that enables developers to score and optimize multimodal AI systems for image-to-text applications.

The new Judge-Image tool, powered by Google Gemini, allows AI engineers to iteratively measure and improve the quality of their multimodal AI applications by scanning for text presence, grid structure, spatial orientation, and object identification.

Also Read: How AI can help Businesses Run Service Centres and Contact Centres at Lower Costs?

“Our mission has always been to advance scalable oversight of AI,” said Anand Kannappan, CEO and Co-founder of Patronus AI. “With the release of GPT-4o, Claude Opus, and Google’s Gemini over the last year, organizations have invested heavily in image generation to drive customer value. However, as these AI experiences scale, so does the unpredictability of LLM systems. Our MLLM-as-a-Judge addresses this critical challenge by providing transparent, reliable evaluation of multimodal systems.”

The Judge-Image tool offers several out-of-box evaluation criteria, including:

• Caption hallucination detection (standard and strict)
• Primary and non-primary object description verification
• Object location accuracy

Beyond validating image caption correctness, Judge-Image can test OCR extraction accuracy for tabular data, AI-generated brand asset accuracy, and scene description validity.

Prior research suggests that Google Gemini can serve as a more reliable MLLM judge compared to alternatives like OpenAI’s GPT-4V, exhibiting less egocentricity and a more equitable approach to judgment. Patronus AI’s internal evaluation datasets confirmed that the Gemini backbone performed better compared to other multimodal LLMs.

Patronus AI plans to expand their multimodal evaluation capabilities to include audio and vision features in future releases.

Latest Read: Taking Generative AI from Proof of Concept to Production

Customer Use Case

Etsy, the leading technology marketplace for independent sellers, has already implemented Patronus AI’s MLLM-as-a-Judge to detect and mitigate caption hallucination from their product images. The Etsy AI team leverages this and the broader Patronus platform to optimize their multimodal AI system.

[To share your insights with us as part of editorial or sponsored content, please write to psen@itechseries.com]