# OmniJev: onejev

> Multimodal decision model in four sizes (0.8B to 27B), fine-tuned from Qwen3.5 and Qwen3.8. Reads a state of screenshots, photos, video frames or text and returns a probability per option for choice, score and noul questions, from option-letter logits in one pass.

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

## Facts

| | |
|---|---|
| Maker | OmniJev (https://systemonemodels.ai/omnijev) |
| Decides | choice, score, noul |
| Architecture | onejev |
| Base model | qwen/qwen3.5-4b |
| Parameters | 4.0B |
| Licence | apache-2.0 |
| Availability | Open weights |
| Released | 2026-09-27 |
| Latest version | 2026.09 |

## Reported evaluation

Suite: OneJev test set (5,701 scored questions, item-disjoint from training; OmniJev's own held-out set). Numbers are the publisher's own.

- Decision accuracy: 75.7%
- Calibration error (ECE): 0.021
- Median latency: 64 ms

## Model card

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

# OneJev-4B

Multimodal decision model in four sizes (0.8B to 27B), fine-tuned from Qwen3.5 and Qwen3.8. Reads a state of screenshots, photos, video frames or text and returns a probability per option for choice, score and noul questions, from option-letter logits in one pass.

Full fine-tunes for one epoch with the vision tower frozen, on 99,193 questions from GUI-agent runs, images, short and long videos, text and rules, with a cross-entropy plus Brier loss. The state is prefilled once and the cache forked per question, so ten questions on one screenshot take 104 ms on an H200. The qev server speaks TypeSafe's System One API plus a `media` field; video needs the PyTorch backend. Siblings: OneJev-0.8B (70.0%), OneJev-9B (76.4%), OneJev-27B on Qwen3.8-27B (77.4%) and OneJev-27B-FP8, on the maker's held-out test set. On DecisionBench (a third-party suite, the maker's own run) the 4B scores 68.9 medium and 45.2 hard, below its untrained base on hard, and the maker's own table shows Jev 1.13 ahead on DecisionBench. 94,707 of the training questions are released; the maker's dataset card says some sources are research-only or non-commercial. PlayJev-0.8B, a separate game-playing policy for 10 browser games, is linked below. Third-party GGUF builds exist and are not listed. By Li Bobo.

## 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 | OmniJev |
| Released | 2026-09-27 |
| License | apache-2.0 |
| Reported accuracy | 75.7% |
| Reported latency | 64 ms for one question, 104 ms for ten questions about one 1280x720 screenshot on one H200 |

## Get the weights

```bash
pip install systemonemodels
systemone pull omnijev/onejev
```

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

## Read more

- [Model card](https://huggingface.co/OmniJev/OneJev-4B)
- [Code](https://github.com/OmniJev/OneJev)
- [Website](https://omnijev.github.io/OneJev/)
- [OneJev-27B](https://huggingface.co/OmniJev/OneJev-27B)
- [OneJev-9B](https://huggingface.co/OmniJev/OneJev-9B)
- [OneJev-0.8B](https://huggingface.co/OmniJev/OneJev-0.8B)
- [Training data](https://huggingface.co/datasets/OmniJev/OneJev-Data)
- [PlayJev-0.8B](https://huggingface.co/OmniJev/PlayJev-0.8B)

---

*This page was opened by System One for OmniJev, 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