# DeskMind: brain

> The decision model of DeskMind, an open-source computer-use agent for macOS: Qwen3.5-4B with a merged LoRA, 8-bit MLX, answering choice, yes/no and score questions from answer-letter logits. Its server speaks /v1/systemone on Apple Silicon; a 0.8B sibling is the fast tier.

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

## Facts

| | |
|---|---|
| Maker | DeskMind (https://systemonemodels.ai/deskmind) |
| Decides | choice, score, noul |
| Architecture | brain |
| Base model | qwen/qwen3.5-4b |
| Parameters | 4.0B |
| Licence | apache-2.0 |
| Availability | Open weights |
| Released | 2026-09-27 |
| Latest version | 0.3.0 |

## Reported evaluation

Suite: JevBench v1.4.2 public set (231 items), maker's local run with the official jevbench.cli (release G18b). Numbers are the publisher's own.

- Decision accuracy: 83.5%
- Calibration error (ECE): 0.089

## Model card

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

# DeskMind Brain 4B

The decision model of DeskMind, an open-source computer-use agent for macOS: Qwen3.5-4B with a merged LoRA, 8-bit MLX, answering choice, yes/no and score questions from answer-letter logits. Its server speaks /v1/systemone on Apple Silicon; a 0.8B sibling is the fast tier.

A personal project of 宗羽 (gxcsoccer). LoRA distillation (KL to teacher distributions plus cross-entropy), merged and quantised to 8-bit MLX, so it runs on Apple Silicon only; keep deskmind.json next to the weights. Option lists longer than 52 are scored in tournament rounds. In the product, brain-0.8b answers first and escalates about 70% of steps at threshold 0.96. The maker's own run of the third-party JevBench v1.4.2 public set gives 0.835 (easy 48/48, original 69/72, hard 76/111; the earlier G14 release scored 0.866) with Brier 0.269 and ECE 0.089; it reports no 13-gram overlap between its training mix and those items. Tuned for macOS Finder and TextEdit, web and form steps; the maker says the probabilities are the model's own weighting, not guarantees. The DeskMind name and logo are excluded from the licence. deskmind/eyes-4b is a GUI-grounding model, not a decision model.

## 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 | DeskMind |
| Released | 2026-09-27 |
| License | apache-2.0 |
| Reported accuracy | 83.5% |

## Get the weights

```bash
pip install systemonemodels
systemone pull deskmind/brain
```

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

## Read more

- [Model card](https://huggingface.co/deskmind/brain-4b)
- [Brain 0.8B (fast tier)](https://huggingface.co/deskmind/brain-0.8b)
- [Code](https://github.com/deskmind-ai/brain)
- [DeskMind](https://deskmind.dev)

---

*This page was opened by System One for DeskMind, 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