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 144M-parameter System One model on the multilingual mmBERT-small encoder. Takes a state, a question and 2 to 20 options and returns one decision with probabilities for choice, score and yes/no questions, on a CPU.
Decides
choice, score, noul
choice, score, noul, classify, route
Architecture
ajev
julia
Fine-tuned from
google/gemma-4-26b-a4b-it
jhu-clsp/mmbert-small
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
—
73.2%
Calibration error
—
—
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 julia-1?
ajev is from Andy Zhang and julia-1 from Supersonic Labs. Both have open weights you can download and run. Both answer choice, score and noul questions. Only julia-1 answers classify and route. julia-1 is the smaller model, at 144M parameters to 26B.
Which is more accurate, ajev or julia-1?
Only julia-1 publishes an accuracy figure (73.2% on typed-decisions test set (400 cases, 2,000 questions)); ajev does not, so there is no comparison to make without your own test.
Which is cheaper, ajev or julia-1?
ajev: Free (open weights). julia-1: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run ajev or julia-1 locally?
Yes, both: systemone pull andy-zhang/ajev and systemone pull supersonic-labs/julia-1 download the weights.
Evaluation suite
Sequential requests, the maker's own measurement
typed-decisions test set (400 cases, 2,000 questions)
