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.
A 0.6B replica of the System One idea trained on game environments. A Qwen3-0.6B backbone with an attention-based Choice head that scores a dynamic candidate set for Maze, Snake, ViZDoom and position prediction, shipped with its full training pipeline.
Decides
choice, score, noul, classify, route
choice, noul, score
Architecture
julia
nanojev
Fine-tuned from
jhu-clsp/mmbert-small
qwen/qwen3-0.6b
License
apache-2.0
mit
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
73.2%
—
Calibration error
—
—
Valid action rate
—
—
Median latency
—
—
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 julia-1 and nanojev?
julia-1 is from Supersonic Labs and nanojev from TianyuCodings. 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 600M. julia-1 is licensed apache-2.0; nanojev, mit.
Which is more accurate, julia-1 or nanojev?
Only julia-1 publishes an accuracy figure (73.2% on typed-decisions test set (400 cases, 2,000 questions)); nanojev does not, so there is no comparison to make without your own test.
Which is cheaper, julia-1 or nanojev?
julia-1: Free (open weights). nanojev: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run julia-1 or nanojev locally?
Yes, both: systemone pull supersonic-labs/julia-1 and systemone pull tianyu-codings/nanojev download the weights.
Evaluation suite
typed-decisions test set (400 cases, 2,000 questions)
