# Everett (everettjf): ezjev

> An independent developer's Jev-style decision model: a LoRA fine-tune of Qwen3.5-4B that answers choice, yes/no and score questions with a probability per option from a single prefill. Weights are open; the maker asks that they be treated as non-commercial.

- Page: https://systemonemodels.ai/everettjf/ezjev
- API: https://api.systemonemodels.ai/v1/models/everettjf/ezjev
- Download: `pip install systemonemodels && systemone pull everettjf/ezjev`

## Facts

| | |
|---|---|
| Maker | Everett (everettjf) (https://systemonemodels.ai/everettjf) |
| Decides | choice, score, noul |
| Architecture | ezjev |
| Base model | qwen/qwen3.5-4b |
| Parameters | 4.0B |
| Licence | No weights licence; maker advises research/non-commercial use (training data includes ANLI, CC BY-NC 4.0); code MIT |
| Availability | Open weights |
| Released | 2026-10-05 |
| Latest version | 2026.09 |

## Reported evaluation

Suite: JevBench public set (231 items; 187 correct), the maker's run of ezjev-4b-s3. Numbers are the publisher's own.

- Decision accuracy: 81.0%
- Calibration error (ECE): 0.044
- Median latency: 34 ms
- p95 latency: 99 ms

## Model card

<!-- generated by scripts/seed_catalog.py; edit content/models/catalog.yaml -->

# ezjev-4b-s3

An independent developer's Jev-style decision model: a LoRA fine-tune of Qwen3.5-4B that answers choice, yes/no and score questions with a probability per option from a single prefill. Weights are open; the maker asks that they be treated as non-commercial.

A merged LoRA; each question takes one prefill and the model's option-letter log-probabilities after "Answer:" give the probabilities, with one global temperature fitted on the maker's held-out dev split. s3 is fine-tuned from s2 (both public). On JevBench's public items (a third-party benchmark) the maker's own run gives 187/231 (easy 48/48, original 71/72, hard 68/111); the maker says no JevBench item was used in training. No licence is declared for the weights and the maker advises research or non-commercial use; the code is MIT.

## What it decides

- **choice** — picks one option from a set
- **score** — places the input on an ordered scale
- **noul** — answers a yes/no question with one probability

## At a glance

| | |
|---|---|
| Parameters | 4B |
| Base model | `qwen/qwen3.5-4b` |
| Maker | Everett (everettjf) |
| Released | 2026-10-05 |
| License | No weights licence; maker advises research/non-commercial use (training data includes ANLI, CC BY-NC 4.0); code MIT |
| Reported accuracy | 81.0% |
| Reported latency | 34 ms p50 / 99 ms p95 on one RTX PRO 6000 |

## Get the weights

```bash
pip install systemonemodels
systemone pull everettjf/ezjev
```

The files are served from the maker's Hugging Face repository, [`everettjf/ezjev-4b-s3`](https://huggingface.co/everettjf/ezjev-4b-s3), and verified against the checksums recorded here.

## Read more

- [Model card (s3)](https://huggingface.co/everettjf/ezjev-4b-s3)
- [ezjev-4b-s2](https://huggingface.co/everettjf/ezjev-4b-s2)
- [Training and eval code](https://github.com/everettjf/ezjev)

---

*This page was opened by System One for Everett (everettjf), who can claim the organisation and take it over at any time.*

---

From System One Models — https://systemonemodels.ai/ · every System One model: https://systemonemodels.ai/system-one-models