# FlyMy.AI: decision-4b

> A LoRA on Qwen3.5-4B that reads option-letter logits in one pass and returns a probability for each declared option of a yes/no, choice or score question, through TypeSafe's /v1/systemone wire format. Adapter and code Apache-2.0; base downloaded separately. Preview.

- Page: https://systemonemodels.ai/flymy-ai/decision-4b
- API: https://api.systemonemodels.ai/v1/models/flymy-ai/decision-4b
- Download: `pip install systemonemodels && systemone pull flymy-ai/decision-4b`

## Facts

| | |
|---|---|
| Maker | FlyMy.AI (https://systemonemodels.ai/flymy-ai) |
| Decides | choice, score, noul |
| Architecture | flymyjev |
| Base model | qwen/qwen3.5-4b |
| Parameters | 4.0B |
| Context | 16K tokens |
| Licence | apache-2.0 |
| Availability | Open weights |
| Released | 2026-09-29 |
| Latest version | 1.5.0 |

## Reported evaluation

Suite: JevBench public set (231 items; 204 correct), maker's package check with the v1.4.2 harness on one RTX 4090; ECE on the hard tier. Numbers are the publisher's own.

- Decision accuracy: 88.3%
- Calibration error (ECE): 0.112
- Median latency: 18.9 ms
- p95 latency: 22 ms

## Model card

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

# Decision 4B v1.5 (FlyMyJev-4B preview)

A LoRA on Qwen3.5-4B that reads option-letter logits in one pass and returns a probability for each declared option of a yes/no, choice or score question, through TypeSafe's /v1/systemone wire format. Adapter and code Apache-2.0; base downloaded separately. Preview.

A rank-16 adapter (57.5 MB); the base downloads separately at a pinned revision. The yes/no temperature (0.40) was chosen so that answers commit, not for calibration (ECE 0.112 on JevBench's public hard tier). Training data is not published; the card says only "our own decision data and licence-reviewed public datasets", and that no JevBench item and no Jev output was used. Needs a CUDA GPU and a C compiler at run time. v1.1 to v1.4 are earlier separate repos (v1.3 is the v1.2 weights with a new yes/no temperature). FlyMy.AI's 2B and Fast previews are evaluation-only and are not listed. Credits on the card: the prompt format is SemIf by TheoLeeCJ (MIT) and the recipe and CUDA-graph design come from JevK5 by allebee. JevBench is a third-party benchmark; the numbers are the maker's own package check. Not affiliated with TypeSafe.

## 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 | FlyMy.AI |
| Released | 2026-09-29 |
| License | apache-2.0 |
| Reported accuracy | 88.3% |
| Reported latency | 18.9 ms p50 / 22.0 ms p95 per short decision on one RTX 4090 with CUDA graphs, in-process; 54 ms eager |

## Get the weights

```bash
pip install systemonemodels
systemone pull flymy-ai/decision-4b
```

The files are served from the maker's Hugging Face repository, [`flymy-ai/decision-4b-v1.5`](https://huggingface.co/flymy-ai/decision-4b-v1.5), and verified against the checksums recorded here.

## Read more

- [Model card](https://huggingface.co/flymy-ai/decision-4b-v1.5)
- [v1.4](https://huggingface.co/flymy-ai/decision-4b-v1.4)
- [JevK5 (recipe credited by the card)](https://huggingface.co/alibiserikbay/JevK5)

---

*This page was opened by System One for FlyMy.AI, 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