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 training-free layer that turns an open LLM into a decision model. It reads typed choice, yes/no and score answers from one prefill, removes option-order bias with no labels and, from a few hundred labels, calibrates or fits a closed-form head. Weights stay untouched.
Decides
choice, score, noul
choice, score, noul, classify, route
Architecture
ajev
anyjev
Fine-tuned from
google/gemma-4-26b-a4b-it
qwen/qwen3-8b
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
—
77.1%
Calibration error
—
0.034
Valid action rate
—
—
Median latency
49 ms
—
p95 latency
—
—
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 anyjev?
ajev is from Andy Zhang and anyjev from Nokia. Both have open weights you can download and run. Both answer choice, score and noul questions. Only anyjev answers classify and route. anyjev is the smaller model, at 8.0B parameters to 26B.
Which is more accurate, ajev or anyjev?
Only anyjev publishes an accuracy figure (77.1% on LocalLLaMA/typed-decisions (2,000 held-out decisions), L2 with 300 labels per question); ajev does not, so there is no comparison to make without your own test.
Which is cheaper, ajev or anyjev?
ajev: Free (open weights). anyjev: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run ajev or anyjev locally?
Yes, both: systemone pull andy-zhang/ajev and systemone pull nokia/anyjev download the weights.
Evaluation suite
Sequential requests, the maker's own measurement
LocalLLaMA/typed-decisions (2,000 held-out decisions), L2 with 300 labels per question
