# Andy Zhang: ajev

> An open Jev-style decision model by Andy Zhang: a LoRA adapter on Gemma 4 26B-A4B (also on Gemma 4 12B) that reads a state and returns a temperature-calibrated probability for every option of a yes/no, choice or score question in one forward pass.

- Page: https://systemonemodels.ai/andy-zhang/ajev
- API: https://api.systemonemodels.ai/v1/models/andy-zhang/ajev
- Download: `pip install systemonemodels && systemone pull andy-zhang/ajev`

## Facts

| | |
|---|---|
| Maker | Andy Zhang (https://systemonemodels.ai/andy-zhang) |
| Decides | choice, score, noul |
| Architecture | ajev |
| Base model | google/gemma-4-26b-a4b-it |
| Parameters | 26B |
| Licence | apache-2.0 |
| Availability | Open weights |
| Released | 2026-10-05 |
| Latest version | 2026.09 |

## Reported evaluation

Suite: Sequential requests, the maker's own measurement. Numbers are the publisher's own.

- Median latency: 49 ms

## Model card

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

# AJev (Gemma 4 26B-A4B, lora1)

An open Jev-style decision model by Andy Zhang: a LoRA adapter on Gemma 4 26B-A4B (also on Gemma 4 12B) that reads a state and returns a temperature-calibrated probability for every option of a yes/no, choice or score question in one forward pass.

The repo holds a rank-32 LoRA (about 149 MB) and per-type temperatures, not the base weights (about 55 GB in bf16). Sibling ajev-gemma4-12b-lora7; the older lora5 is kept for leaderboard reproduction. The maker's own held-out runs: 0.887 on JevBench's public items (a third-party benchmark), 0.790 on Kev's transfer set, 0.780 on typed-decisions. His own run of the third-party Decision Index 0.2.1 gives 57.42 (not reproduced by the maintainers; the board lists lora5 at 50.02). Trained in part on training splits of benchmarks related to that board, with decontamination stated. The cards say some training data carries non-commercial or attribution terms. Inference code (cmzy/ajev-infer, Apache-2.0) serves /v1/systemone. Not affiliated with TypeSafe AI.

## 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 | 26B |
| Base model | `google/gemma-4-26B-A4B-it` |
| Maker | Andy Zhang |
| Released | 2026-10-05 |
| License | apache-2.0 |
| Reported latency | 49 ms median for sequential requests on one RTX PRO 6000 |

## Get the weights

```bash
pip install systemonemodels
systemone pull andy-zhang/ajev
```

The files are served from the maker's Hugging Face repository, [`andyzhang232/ajev-gemma4-26b-a4b-lora1`](https://huggingface.co/andyzhang232/ajev-gemma4-26b-a4b-lora1), and verified against the checksums recorded here.

## Read more

- [Model card](https://huggingface.co/andyzhang232/ajev-gemma4-26b-a4b-lora1)
- [AJev 12B (lora7)](https://huggingface.co/andyzhang232/ajev-gemma4-12b-lora7)
- [Inference code](https://github.com/cmzy/ajev-infer)

---

*This page was opened by System One for Andy Zhang, 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