title: "Right-sizing intelligence: Glean saves 81% on token costs and is preferred 78% of the time versus Claude Cowork"

Choosing the right AI model for a given task is crucial for balancing cost, speed, and quality in an enterprise setting. Glean's platform utilizes an auto-routing system that dynamically selects the most appropriate model from a hub of options, including both large frontier models and smaller, faster ones. In a benchmark of over 180 enterprise tasks, this auto-routing approach was compared against using a single powerful model, Claude 3.5 Sonnet. The results showed that Glean's method was preferred by human evaluators 78% of the time while also reducing token costs by 81%, demonstrating that intelligent routing can deliver better, more cost-effective results.