# GeneLab: sokudan-ja-310m

> Japanese decision model by GeneLab, an individual developer in Tokyo: a 315M ModernBERT-ja encoder that answers choice, score and yes/no (noul) questions about a text, with a probability per option, in one forward pass per question. Only yes/no is calibrated.

- Page: https://systemonemodels.ai/genelab/sokudan-ja-310m
- API: https://api.systemonemodels.ai/v1/models/genelab/sokudan-ja-310m
- Download: `pip install systemonemodels && systemone pull genelab/sokudan-ja-310m`

## Facts

| | |
|---|---|
| Maker | GeneLab (https://systemonemodels.ai/genelab) |
| Decides | choice, score, noul |
| Architecture | sokudan |
| Base model | sbintuitions/modernbert-ja-310m |
| Parameters | 315M |
| Licence | apache-2.0 |
| Availability | Open weights |
| Released | 2026-09-29 |
| Latest version | 0.2.1 |

## Reported evaluation

Suite: bench_ja (300 synthetic Japanese business inquiries, GeneLab's own set): choice accuracy of v0.2; ECE is yes/no after the v0.2.1 temperature. Numbers are the publisher's own.

- Decision accuracy: 88.0%
- Calibration error (ECE): 0.105
- Median latency: 22.8 ms
- p95 latency: 27.5 ms

## Model card

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

# sokudan-ja-310m

Japanese decision model by GeneLab, an individual developer in Tokyo: a 315M ModernBERT-ja encoder that answers choice, score and yes/no (noul) questions about a text, with a probability per option, in one forward pass per question. Only yes/no is calibrated.

Japanese only; the maker says not to use the yes/no type on English input. Choice and score are not calibrated, and yes/no underpredicts true and needs its own threshold. On the maker's own bench_ja (300 synthetic Japanese business inquiries) choice accuracy is 0.880, score accuracy 0.817 and yes/no accuracy 0.780 (AUROC 0.844). Accuracy drops on states longer than about 400 tokens; training data and benchmarks are both synthetic, from one generator model; the maker says not to use it for hiring, credit, medical or legal decisions. Needs the sokudan package (custom scorer and ordinal heads); `sokudan serve` offers a /v1/systemone-compatible server, and an MLX path runs on Apple Silicon. The maker did not measure TypeSafe Jev, citing Jev's terms.

## 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 | 315M |
| Base model | `sbintuitions/modernbert-ja-310m` |
| Maker | GeneLab |
| Released | 2026-09-29 |
| License | apache-2.0 |
| Reported accuracy | 88.0% |
| Reported latency | 22.8 ms p50 / 27.5 ms p95 per three-question request on an RTX 5090; 371 ms on a Core Ultra 9 285K CPU; about 10 ms per short question with MLX on an M1 Max |

## Get the weights

```bash
pip install systemonemodels
systemone pull genelab/sokudan-ja-310m
```

The files are served from the maker's Hugging Face repository, [`GeneLab/sokudan-ja-310m`](https://huggingface.co/GeneLab/sokudan-ja-310m), and verified against the checksums recorded here.

## Read more

- [Model card](https://huggingface.co/GeneLab/sokudan-ja-310m)
- [Code](https://github.com/hiroki-abe-58/sokudan)
- [PyPI package](https://pypi.org/project/sokudan/)

---

*This page was opened by System One for GeneLab, 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