# New decision model Jev matches many LLM judges in pediatric AI evaluation

> Researchers report Jev, a cheaper decision language model, agrees with physician ratings as well as several top‑tier LLM judges on clinical tests.

Oossa · 2026-10-22 · https://oossa.com/en/new-decision-model-jev-matches-many-llm-judges-in-pediatric-ai-evaluation

A team of Israeli researchers introduced Jev, a decision language model that gives structured judgments on clinical AI output. They tested it on four tasks, including 118 pediatric answers graded by doctors. Jev’s agreement score (Gwet's AC2 0.50) was on par with or better than seven of the 12 large language model (LLM) judges they compared, such as Claude Opus 5.5 (0.49) and GPT‑6 (0.38). Jev also matched or beat LLM judges on binary HealthBench criteria and performed similarly to a three‑LLM jury on MedHELM tasks. The paper notes Jev costs 73‑294× less and runs 7‑57× faster than the frontier LLM judges.

## The facts

- Jev’s Gwet's AC2 agreement with physicians: 0.50
- Frontier LLM judges cost 73‑294 times more and take 7‑57 times longer than Jev

## Why it matters

Hospitals could evaluate AI tools for children more affordably and quickly using Jev instead of expensive, slow LLM judges.

## Sources & references

1. [Decision language models as affordable and rapid alternatives to LLM judges of clinical generative AI](https://arxiv.org/abs/2610.28826) – arXiv, 2026-10-22

Last updated: 2026-10-22