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

A benchmark of over 180 enterprise tasks demonstrates that "right-sizing" AI through auto-routing can significantly reduce costs while improving performance. Glean's platform, which routes tasks to the most appropriate model, achieved an 81% savings on token costs. In a direct comparison against Claude Cowork, Glean's approach was also preferred by users 78% of the time. This highlights the effectiveness of using high-quality context and intelligent model selection for enterprise AI applications.