Mysterious Hy3 LLM unexpectedly dominates OpenRouter model rankings
By
freediver
3d ago· 9 min readenInsight
100/100
Golden Brown
Bagelometer↗
A baker's-dozen of insight crammed into one ring.
Score100TypeanalysisSentimentneutral
Summary
An analysis of the mysterious "Hy3" LLM that has unexpectedly topped OpenRouter's AI Model Rankings by a significant margin. The article explores what Hy3 is, its origins, performance characteristics, and the implications of an unknown model outperforming established players like GPT-4 and Claude on real-world user interaction data. The piece examines OpenRouter's role as an intermediary providing transparent usage data, and questions what Hy3's sudden rise means for the AI landscape.
Key quotes
· 3 pulledOpenRouter has robust, representative data on how users interact with LLMs and it publishes this data on the AI Model Rankings page: a welcome deviation from the labs themselves which generally keep this data secret for competitive reasons.
Recently, I checked the OpenRouter rankings and noticed something peculiar.
Two new model
Why Hy3?
