Jev in production › Model and agent routing
Jevstiller distills a repeated Jev classification call into a local model over time; a demo shows the local model taking over after about 4,000 requests (tomerglick57).

Jev, distilled on the fly. Same call. Same answers. Your hardware.
Put Jevstiller in front of a repeated Jev classification call. At first every request still goes to Jev. From Jev's own answers — with their full probability distributions — it trains a small local model on your traffic, checks that the model agrees with Jev within a budget you set, and then answers most requests itself. Uncertain or novel input, and a permanent random audit slice, keep going to Jev. Every request answered locally is a Jev call you don't pay for, one that can't hit a rate limit, and an answer in ~15 ms instead of ~300 ms.
For the project's own README, linking back here: