# Valen Team: valen

> A multimodal decision model from the Valen team: Qwen3.5-4B plus a two-layer MLP-Mixer decision head that returns probability distributions for Choice, Noul and Score questions about a shared state of text, images or video, without generating answer text.

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

## Facts

| | |
|---|---|
| Maker | Valen Team (https://systemonemodels.ai/valen-team) |
| Decides | choice, score, noul |
| Architecture | valen |
| Base model | qwen/qwen3.5-4b |
| Parameters | 4.5B |
| Licence | apache-2.0 |
| Availability | Open weights |
| Released | 2026-10-06 |
| Latest version | 2026.09 |

## Reported evaluation

Suite: VisualDecisionBench-Image (the Valen team's own benchmark, 2,000 questions over 1,402 image states). Numbers are the publisher's own.

- Decision accuracy: 83.1%

## Model card

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

# Valen-4B

A multimodal decision model from the Valen team: Qwen3.5-4B plus a two-layer MLP-Mixer decision head that returns probability distributions for Choice, Noul and Score questions about a shared state of text, images or video, without generating answer text.

Trained in two SFT stages (decision-head warm-up, then joint LoRA, mixer and visual-merger training); videos are sampled at 16 frames. Several questions can share one state or be scored separately. Loading needs trust_remote_code. On its own VisualDecisionBench (released alongside) the maker reports 83.06% image and 87.36% video accuracy for the 4B; its own run on JevBench's public items (a third-party benchmark) gives 83.12%. The card notes that the released merged weights can differ slightly in probability from the training checkpoints behind those numbers; no calibration figures are published. Siblings Valen-2B and 0.8B, and an earlier Valen-Preview-0923. Base models and source datasets keep their own licences.

## 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 | 4.5B |
| Base model | `qwen/qwen3.5-4b` |
| Maker | Valen Team |
| Released | 2026-10-06 |
| License | apache-2.0 |
| Reported accuracy | 83.1% |

## Get the weights

```bash
pip install systemonemodels
systemone pull valen-team/valen
```

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

## Read more

- [Model card](https://huggingface.co/Valen-Team/Valen-4B)
- [Valen-2B](https://huggingface.co/Valen-Team/Valen-2B)
- [Code](https://github.com/Liuziyu77/Valen)

---

*This page was opened by System One for Valen Team, 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