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.
A 4B decision model fine-tuned from Qwen3.5-4B-Base. A state and typed questions (choice, yes/no, score) go in; a temperature-calibrated probability for every option comes out of one pass, through a self-hosted /v1/systemone-compatible server. Apache-2.0 weights and code.
Decides
choice, score, noul
choice, score, noul, classify, route
Architecture
ajev
wald
Fine-tuned from
google/gemma-4-26b-a4b-it
qwen/qwen3.5-4b-base
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
—
88.3%
Calibration error
—
0.045
Valid action rate
—
—
Median latency
49 ms
33 ms
p95 latency
—
168 ms
Evaluation suite
Figures are from each model’s manifest; accuracy and latency are what the publishers report, on their own suites and hardware. Add a third model.
Questions
What is the difference between ajev and wald?
ajev is from Andy Zhang and wald from ORG2 AI. Both have open weights you can download and run. Both answer choice, score and noul questions. Only wald answers classify and route. wald is the smaller model, at 4.2B parameters to 26B.
Which is more accurate, ajev or wald?
Only wald publishes an accuracy figure (88.3% on JevBench public set (231 items; 204 correct), maker's run with the JevBench CLI, v1.2, effort none; items used as a development scoreboard); ajev does not, so there is no comparison to make without your own test.
Which is cheaper, ajev or wald?
ajev: Free (open weights). wald: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Which is faster, ajev or wald?
By their publishers’ figures, wald answers in about 33 ms at the median and ajev in about 49 ms — measured on different hardware, so treat it as a rough guide.
Can I run ajev or wald locally?
Yes, both: systemone pull andy-zhang/ajev and systemone pull org2ai/wald download the weights.
Sequential requests, the maker's own measurement
JevBench public set (231 items; 204 correct), maker's run with the JevBench CLI, v1.2, effort none; items used as a development scoreboard
Latest version
2026.09
1.2.0
Variants
—
Dockerfile, LICENSE, NOTICE, model-files, reference, server
