# Rizzo AI Academy: rizzo-flow

> A local, Jev-compatible decision model from Rizzo AI Academy. XHToken's Spark-X2.5-4B with a merged typed-decisions LoRA, run on llama.cpp; yes/no, choice and score questions share one prefill of the state and are read from the answer-letter logits. No text is generated.

- Page: https://systemonemodels.tech/rizzo-ai-academy/rizzo-flow
- API: https://api.systemonemodels.tech/v1/models/rizzo-ai-academy/rizzo-flow
- Download: `pip install systemonemodels && systemone pull rizzo-ai-academy/rizzo-flow`

## Facts

| | |
|---|---|
| Maker | Rizzo AI Academy (https://systemonemodels.tech/rizzo-ai-academy) |
| Decides | choice, score, noul, classify, route |
| Architecture | rizzo-flow |
| Base model | xhtoken/spark-x2.5-4b |
| Parameters | 4.0B |
| Licence | apache-2.0 |
| Availability | Open weights |
| Released | 2026-09-25 |
| Latest version | 2026.09 |

## Reported evaluation

Suite: LocalLLaMA/typed-decisions test split (400 cases, 2,000 decisions), Q8_0. Numbers are the publisher's own.

- Decision accuracy: 64.8%
- Calibration error (ECE): 0.112
- Median latency: 195 ms

## Model card

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

# Rizzo Flow 4B

A local, Jev-compatible decision model from Rizzo AI Academy. XHToken's Spark-X2.5-4B with a merged typed-decisions LoRA, run on llama.cpp; yes/no, choice and score questions share one prefill of the state and are read from the answer-letter logits. No text is generated.

On the test split of LocalLLaMA/typed-decisions the Q8_0 build scores 64.8% with a calibration error of 0.112, against 57.4% (0.349) for the original Spark-X2.5-4B on the same RTX 5060 Ti; Jev's published figure there is 72.7%. The authors say the probabilities are not calibrated for a new domain and ship temperature scaling for your own data. The rank-16 LoRA was trained for one epoch on 28,321 public questions, none labelled by Jev. The repository holds the merged model as BF16 safetensors and as Q8_0, Q4_K_M and BF16 GGUF, plus the adapter. The server also offers /v1/decisions, with numeric answers and abstention; choice takes up to 26 options, and a question 8,192 tokens by default (1M native). The runtime has been public since 21 September on the original weights; this fine-tune followed on the 25th. Sibling: rizzo-flow-1.7b (54.4%). Some training configs are CC-BY-SA or CC-BY-4.0; check them before redistributing derivatives.

## 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 calibrated probability
- **classify** — assigns a category from a fixed taxonomy
- **route** — sends the input to one of several destinations

## At a glance

| | |
|---|---|
| Parameters | 4B |
| Base model | `xhtoken/spark-x2.5-4b` |
| Maker | Rizzo AI Academy |
| Released | 2026-09-25 |
| License | apache-2.0 |
| Reported accuracy | 64.8% |
| Reported latency | 195 ms p50 per five-question case on an RTX 5060 Ti (Q8_0); about 50 ms for one short decision |

## Get the weights

```bash
pip install systemonemodels
systemone pull rizzo-ai-academy/rizzo-flow
```

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

## Read more

- [Source and runtime](https://github.com/Rizzo-AI-Academy/rizzo-flow)
- [Weights on Hugging Face](https://huggingface.co/rizzoaiacademy/rizzo-flow)
- [rizzo-flow-1.7b](https://huggingface.co/rizzoaiacademy/rizzo-flow-1.7b)
- [Project website](https://rizzo-ai-academy.github.io/rizzo-flow/)

---

*This page was opened by System One for Rizzo AI Academy, who can claim the organisation and take it over at any time.*

---

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