# Oraculum: manchego

> Oraculum's 4B typed-decision model: a merged LoRA on Qwen3.5-4B that reads a state, a question and lettered options and returns a probability for each option (choice, yes/no noul, ordered score) in one forward pass. Apache-2.0, with MLX builds and an offline System One server.

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

## Facts

| | |
|---|---|
| Maker | Oraculum (https://systemonemodels.ai/oraculum) |
| Decides | choice, score, noul |
| Architecture | manchego |
| Base model | qwen/qwen3.5-4b |
| Parameters | 4.0B |
| Licence | apache-2.0 |
| Availability | Open weights |
| Released | 2026-09-30 |
| Latest version | 3.0.0 |

## Reported evaluation

Suite: JevBench public set (231 decisions), the maker's run, not an official JevBench score. Numbers are the publisher's own.

- Decision accuracy: 85.7%
- Calibration error (ECE): 0.055

## Model card

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

# Manchego v3

Oraculum's 4B typed-decision model: a merged LoRA on Qwen3.5-4B that reads a state, a question and lettered options and returns a probability for each option (choice, yes/no noul, ordered score) in one forward pass. Apache-2.0, with MLX builds and an offline System One server.

v3 is a rank-16 LoRA, merged, trained under SemIf's direct-options prompt (credited, MIT); v2.1 stays at its own tag. Probabilities are a softmax over the option-letter logits. The maker's numbers are labelled "not an official JevBench score". It is overconfident on unfamiliar tasks (ECE 0.129 on 27 held-out Natural Instructions tasks), and the serving temperature map makes served yes/no probabilities sharp. MLX 8-bit and 4-bit builds; the serving repo speaks /v1/systemone offline.

## 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 | Oraculum |
| Released | 2026-09-30 |
| License | apache-2.0 |
| Reported accuracy | 85.7% |

## Get the weights

```bash
pip install systemonemodels
systemone pull oraculum/manchego
```

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

## Read more

- [Model card](https://huggingface.co/oraculumai/Manchego)
- [Serving code](https://github.com/nschlaepfer/manchego-serve)

---

*This page was opened by System One for Oraculum, 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