Research explainer · XiaoHu reads the paper

Repeated one-token answers can flag a model endpoint mismatch

A study of 165 served models found a proprietary-branded endpoint statistically indistinguishable from a Qwen deployment under its fingerprint test. This is a signal to investigate, not proof of substitution.
The problem

You pay for a name, not for a machine

You pay for a top-tier model; the thing answering you might be something much cheaper. Aggregators like OpenRouter are genuinely convenient — hundreds of models behind one endpoint. But all you select is a name. Which machine your request lands on, and what's running there, is invisible to you, and nothing certifies it.

Primary source and reference