Show HN: Jevstiller – Distill Jev into a local model, with a disagreement bound

Discussion (4 Comments)Read Original on HackerNews

gingersnap•about 2 hours ago

Is the local model similar to model2vec?

tgluck•about 1 hour ago

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

tgluck•about 5 hours ago

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.

wedg_•20 minutes ago

Woah cool idea. So it's almost a drop-in replacement for a typical Jev setup that just reduces your jev bill over time ?

Show HN: Jevstiller – Distill Jev into a local model, with a disagreement bound

⚡ Community Insights

Discussion (4 Comments)Read Original on HackerNews