Voice AI models and benchmarks
Every model Patchboard has measured, and the 72 of 72 catalog entries a request can pin today. Accuracy, latency and price as each board published them — nothing on this page is estimated to fill a column.
22 studies across 11 languages
Naturalness and error rate are measured per language, not inferred from English — the model that wins one study routinely loses another.
What benchmarks.patchboard.online picks for each job
The measurement that decided it. A stack whose legs are identical to another's is one row: the publisher distinguishes four jobs and its boards distinguish two stacks. A stack with a leg Router cannot currently call is not shown at all.
Reading the boards correctly
An id offered by the current Router catalog can be sent as model; other ids identify benchmark evidence without claiming that the route is currently available. A direct benchmark label means the provider published no Router model id for that row.
A dash in a column of word error rates reads as a perfect score, so an unmeasured cell says so in words. A leading ~ is the board's own mark for an estimate.
The # column indexes the order you sorted into. It is not a verdict.