Cloudflare Auto Router picks cheaper models per request in AI Gateway
Source
cfl.re
Author
Cloudflare
Date
Why it matters
Auto Router sends each request to a cheaper or stronger model based on a complexity classifier, so apps can cut tokenThe chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.Full definition → spend without hand-written routing rules. It is in public beta through AI Gateway.
Terms in this piece · Glossary
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
postCloudflare AI Search goes GA with image embeddings and PDF OCR
articleWe are introducing Clef and Clef-flash, open-source decision models hosted on…
articleToday, we’re making the Cloudflare Monetization Gateway available as part of a…
articleAI Gateway User Insights flags model overuse by task, model and user
