Case Study: Yext achieves 2x faster time-to-market and 80% lower compute costs with BentoML

Yext cuts compute costs by 80% with BentoML

Yext, a leading digital presence platform, faced challenges scaling its AI workloads due to rigid infrastructure, time-consuming model production, and cross-team dependencies. This slowed their deployment speed and innovation. They selected BentoML's unified inference platform, specifically utilizing BentoCloud with its BYOC option, to address these issues.

By implementing BentoML, Yext standardized its development process, enabling its Data Science team to self-serve deployments. This resulted in a 70% reduction in development time and allowed the company to ship twice as many models. Furthermore, BentoML's multi-cloud, multi-region deployment and efficient autoscaling capability helped Yext reduce its compute costs by up to 80%, all while running over 150 models in production.

Yext