# Shanghua Gao: rsi-jev

> Jev-style decision model fine-tuned from Qwen3.5-4B-Base by a self-improving loop of AI agents. Answers noul, choice and score questions about text and up to four images in one pass, with a calibrated probability per option; effort picks an exit at layer 16, 20 or 32.

- Page: https://systemonemodels.ai/shanghua-gao/rsi-jev
- API: https://api.systemonemodels.ai/v1/models/shanghua-gao/rsi-jev
- Download: `pip install systemonemodels && systemone pull shanghua-gao/rsi-jev`

## Facts

| | |
|---|---|
| Maker | Shanghua Gao (https://systemonemodels.ai/shanghua-gao) |
| Decides | choice, score, noul |
| Architecture | rsi-jev |
| Base model | qwen/qwen3.5-4b-base |
| Parameters | 4.7B |
| Context | 32K tokens |
| Licence | apache-2.0 |
| Availability | Open weights |
| Released | 2026-10-06 |
| Latest version | 6.0.0 |

## Reported evaluation

Suite: RSI-Jev held-out set eval_final_v2 (the maker's zero-shot check that no release trained on), as served by default. Numbers are the publisher's own.

- Decision accuracy: 69.8%
- Calibration error (ECE): 0.036

## Model card

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

# RSI-Jev v6.0-VL 4B

Jev-style decision model fine-tuned from Qwen3.5-4B-Base by a self-improving loop of AI agents. Answers noul, choice and score questions about text and up to four images in one pass, with a calibrated probability per option; effort picks an exit at layer 16, 20 or 32.

A personal project of Shanghua Gao. An agent loop trained, evaluated and documented each release: seven releases in twelve days, with hundreds of experiments logged, failures included. No generated text; `effort` low, medium, high or auto picks the exit depth, and image questions always use all 32 layers. Ships a Jev-compatible HTTP server and a pip package. Training data includes benchmark train splits; the maker found about 1,000 Decision Index test items in his corpora and re-scored without them, and his own run of that third-party board gives 46.24. His internal 15-benchmark suite (0.770) overlaps training data, so the held-out number below is the cleaner one. The weights are Apache-2.0, but the vision releases were trained on several non-commercial or research-only image datasets. Siblings: v5.0-VL 3B, v4.0-VL 2B and earlier text-only releases.

## 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 | 4.69B |
| Base model | `Qwen/Qwen3.5-4B-Base` |
| Maker | Shanghua Gao |
| Released | 2026-10-06 |
| License | apache-2.0 |
| Reported accuracy | 69.8% |

## Get the weights

```bash
pip install systemonemodels
systemone pull shanghua-gao/rsi-jev
```

The files are served from the maker's Hugging Face repository, [`shgao/rsi-jev-v6.0-vl-4b`](https://huggingface.co/shgao/rsi-jev-v6.0-vl-4b), and verified against the checksums recorded here.

## Read more

- [Model card](https://huggingface.co/shgao/rsi-jev-v6.0-vl-4b)
- [Code and experiment log](https://github.com/Shanghua-Gao/RSI-Jev)
- [RSI-Jev v5.0-VL 3B](https://huggingface.co/shgao/rsi-jev-v5.0-vl-3b)

---

*This page was opened by System One for Shanghua Gao, 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