# AutoTrust AI Lab: jev-27b

> AutoTrust's student of TypeSafe Jev 1.13. A frozen Qwen3.8-27B plus a 108.9M-parameter decision block trained on Jev's own output distributions; one set of weights answers typed questions in one pass (System 1) or generates text with the untouched base (System 2).

- Page: https://systemonemodels.tech/autotrust-ai/jev-27b
- API: https://api.systemonemodels.tech/v1/models/autotrust-ai/jev-27b
- Download: `pip install systemonemodels && systemone pull autotrust-ai/jev-27b`

## Facts

| | |
|---|---|
| Maker | AutoTrust AI Lab (https://systemonemodels.tech/autotrust-ai) |
| Decides | choice, score, noul |
| Architecture | blocks-of-experts |
| Base model | qwen/qwen3.8-27b |
| Parameters | 27B |
| Licence | apache-2.0 |
| Availability | Open weights |
| Released | 2026-09-25 |
| Latest version | 2026.09 |

## Reported evaluation

Suite: JevBench public set (231 items), family-macro score, maker's run. Numbers are the publisher's own.

- Decision accuracy: 88.7%

- Median latency: 137 ms

## Model card

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

# JEV-27B

AutoTrust's student of TypeSafe Jev 1.13. A frozen Qwen3.8-27B plus a 108.9M-parameter decision block trained on Jev's own output distributions; one set of weights answers typed questions in one pass (System 1) or generates text with the untouched base (System 2).

Noul, Choice over 2 to 16 options, and Score on a 0 to 5 scale, served with vLLM. On a held-out set labelled mostly by Jev 1.13 itself, mean KL to Jev is about 0.017 and ECE 0.0009; the maker's run of JevBench's 231 public items gives a family-macro 88.7 against 87.2 for its own run of Jev; median 137 ms per decision on one B200. HumanEval stays at 78.0% because the base is not modified. Trained from the Apache-2.0 corpus SargeDev/jev-distill-corpus-v3, i.e. distilled from Jev's outputs; AutoTrust states no affiliation with TypeSafe. Check TypeSafe's terms before mirroring or recommending it. Siblings: JEV-9B (Qwen3.5-9B), JEV-Gemma4-26B-A4B and JEV-27B-VL.

## 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

## At a glance

| | |
|---|---|
| Parameters | 27B |
| Base model | `qwen/qwen3.8-27b` |
| Maker | AutoTrust AI Lab |
| Released | 2026-09-25 |
| License | apache-2.0 |
| Reported accuracy | 88.7% |
| Reported latency | 137 ms median per single decision on one B200; 4.2 ms per decision in vLLM batches of 128 |

## Get the weights

```bash
pip install systemonemodels
systemone pull autotrust-ai/jev-27b
```

The files are served from the maker's Hugging Face repository, [`autotrust/JEV-27B`](https://huggingface.co/autotrust/JEV-27B), and verified against the checksums recorded here.

## Read more

- [Model card](https://huggingface.co/autotrust/JEV-27B)
- [JEV-9B](https://huggingface.co/autotrust/JEV-9B)
- [Training corpus](https://huggingface.co/datasets/SargeDev/jev-distill-corpus-v3)

---

*This page was opened by System One for AutoTrust AI Lab, 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