Cloudflare open-sourced small models that can only choose from a list you give them
On 1st Oct'26, Cloudflare released Clef and Clef-flash, two open-weight models under the Apache 2.0 licence, built on Qwen models of 27B and 9B parameters. You give them a schema, such as a set of categories or a yes-or-no question, and they return a typed answer with a probability attached rather than free text. Cloudflare reports a median response time of about 39 milliseconds for Clef-flash, and the models run on its Workers AI platform or can be downloaded from Hugging Face. Fine-tuning is offered through Cloudflare's own engineers for now, with a self-serve option promised later.
What it actually means
Most agent pipelines pay a frontier model's price to answer questions with a handful of possible answers, such as which team gets this ticket or whether a message needs a human. A model that can only pick from your list cannot wander off into prose, and the probability it returns gives you a threshold to route on, so the confident cases flow through and the uncertain ones go to a person. That is the human gate most teams describe and few build, because a free-text answer gives you nothing to set the threshold against.
The latency figures are Cloudflare's own, measured on its own tasks, so test Clef-flash against your current model on a few hundred of your real tickets before switching anything. If it holds, the routing step in your pipeline gets faster, and the expensive model is left for the work that needs it.
Stop paying a frontier model to answer a multiple-choice question.
Source: Cloudflare · Tuesday 6 October 2026