← Tuesday, September 29, 2026Models & trainingTool

Jeff decision models return calibrated class probabilities in one forward pass

github.com 8 min read Editor score 8/10

The small fine-tunes target zero-shot classification and take about 22 ms per decision on an RTX PRO 6000 or 28 ms on an Apple M4 Max. A voice-navigation fine-tune raised held-out accuracy from 31.7% to 95.8% in under half an hour on one GPU.

What

Jeff is a family of small, fine-tuned models for zero-shot classification that return calibrated probabilities for listed options.

How

A single forward pass produces probabilities, a selected option, and confidence, without generating text or parsing a response.

Why it matters

Its compact decision interface offers engineers a way to add fast classification to local applications.

Read on github.com → Hacker News · 170 comments

Headline, summary, brief and diagram are written by a model from the source's own text; the link, byline and numbers come from the source. Read the original before relying on a claim.