Potential Savings

Assess

Tools

A view that identifies requests that may be served by faster or cheaper models without hurting output quality.

Why it's here

Placed in Assess: 1 article(s) of evidence from 1 source(s), led by product launches, with 1 in the last 30 days. Confidence 24%. Low accumulated evidence, so it defaults conservatively pending more signal.

Evidence (1)

6Cloudflare Blog·9/30/2026product_launch
Cloudflare adds AI model overuse insights to User Insights
Cloudflare updated User Insights in AI Gateway to show when selected AI models may be more capable than a task requires. The feature helps teams analyze usage by user, agent, application, task, cost, latency, and conversation patterns to identify inefficient or excessive model use, and it is available free to AI Gateway users.