Obviously it only helps when the same question is asked many times, but that's the case it's built for, and the case I have, every Jev question in my other project repeats thousands of times, and most Jev uses I've seen online look the same
Author here. This puts a proxy in front of repeated Jev classification calls. At first everything goes to Jev; from Jev's answers it trains a small head on frozen sentence embeddings, picks a confidence threshold with an exact finite-sample bound so that at most 2% of all requests get an answer Jev wouldn't have given, and then answers the confident share locally at ~15 ms on a CPU. A permanent 2% audit keeps checking; if agreement breaks, everything falls back to Jev and it retrains.
Known limits: agreement is not accuracy (if Jev is wrong, so is the local model); coverage tracks how consistent Jev itself is (22% on noisy tweet tasks, 80% on news); it speaks Jev's API only, an OpenAI-compatible front is on the roadmap. Since 0.4.0 the guarantee can also cover "would Jev have been unsure", which matters if your code routes low-confidence answers to review. Apache 2.0.
Not today, but Interesting idea. The main motivation was a drop-in for an existing Jev setup, so the only teacher right now is Jev and the audit measures agreement with Jev. A correction would have to become a second label source that overrides Jev's for that input.
The head is a multinomial logistic regression: one linear layer plus softmax on top of a frozen sentence-embedding model (bge-small by default, swappable). That head is the entire local model, the encoder is off the shelf and never changes.
This will only work for simple text classification tasks, which is the least interesting possible use of Jev.
Obviously it only helps when the same question is asked many times, but that's the case it's built for, and the case I have, every Jev question in my other project repeats thousands of times, and most Jev uses I've seen online look the same
Is the local model similar to model2vec?
Not really
they distill different things. model2vec distills a sentence transformer into static embeddings, so the output is a faster general-purpose encoder.
Jevstiller keeps the encoder frozen (bge-small by default) and distills Jev's decisions on one specific question into a small head on top of it
Author here. This puts a proxy in front of repeated Jev classification calls. At first everything goes to Jev; from Jev's answers it trains a small head on frozen sentence embeddings, picks a confidence threshold with an exact finite-sample bound so that at most 2% of all requests get an answer Jev wouldn't have given, and then answers the confident share locally at ~15 ms on a CPU. A permanent 2% audit keeps checking; if agreement breaks, everything falls back to Jev and it retrains.
Known limits: agreement is not accuracy (if Jev is wrong, so is the local model); coverage tracks how consistent Jev itself is (22% on noisy tweet tasks, 80% on news); it speaks Jev's API only, an OpenAI-compatible front is on the roadmap. Since 0.4.0 the guarantee can also cover "would Jev have been unsure", which matters if your code routes low-confidence answers to review. Apache 2.0.
Isn't this against their ToS? Useful for hobby stuff.
It would be great if we could correct Jev's incorrect answers, even on a separate endpoint. Let me tell it what Jev got wrong.
What type of head is that? What type of model is that head part of?
Not today, but Interesting idea. The main motivation was a drop-in for an existing Jev setup, so the only teacher right now is Jev and the audit measures agreement with Jev. A correction would have to become a second label source that overrides Jev's for that input.
The head is a multinomial logistic regression: one linear layer plus softmax on top of a frozen sentence-embedding model (bge-small by default, swappable). That head is the entire local model, the encoder is off the shelf and never changes.
Yeah, I kinda figured that head was the whole model, the way you phrased it. scikit-learn?
No, plain numpy. It's full-batch Adam on cross-entropy against soft targets, about 80 lines.
I'll look into Adam, thank you!
Woah cool idea. So it's almost a drop-in replacement for a typical Jev setup that just reduces your jev bill over time ?
Thanks.
Drop-in yes: point TYPESAFE_BASE_URL at it and nothing else changes.
Please don’t post AI generated replies.
Why do you think that's AI?