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 364M Hebrew decision encoder on NeoDictaBERT: give it a state, a question and 2 to 7 free-text options and it returns a probability for each option in one forward pass, about 70 ms on a laptop CPU. Hebrew only.
Decides
choice, score, noul, classify, route
choice, classify, route
Architecture
julia
nativ
Fine-tuned from
jhu-clsp/mmbert-small
dicta-il/neodictabert
License
apache-2.0
cc-by-4.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
73.2%
96.2%
Calibration error
—
0.010
Valid action rate
—
—
Median latency
—
—
p95 latency
—
—
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 julia-1 and nativ-he-decision?
julia-1 is from Supersonic Labs and nativ-he-decision from Yoav Pinto. Both have open weights you can download and run. Both answer choice, classify and route questions. Only julia-1 answers score and noul. julia-1 reads up to 8K tokens of state, against 1K tokens for nativ-he-decision. julia-1 is the smaller model, at 144M parameters to 364M. julia-1 is licensed apache-2.0; nativ-he-decision, cc-by-4.0.
Which is more accurate, julia-1 or nativ-he-decision?
They report on different suites — julia-1 73.2% on typed-decisions test set (400 cases, 2,000 questions), nativ-he-decision 96.2% on nativ-bench user-intent task, 4 options (MASSIVE Hebrew, 2,973 items; the maker's own suite) — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, julia-1 or nativ-he-decision?
julia-1: Free (open weights). nativ-he-decision: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run julia-1 or nativ-he-decision locally?
Yes, both: systemone pull supersonic-labs/julia-1 and systemone pull yoav-pinto/nativ-he-decision download the weights.
typed-decisions test set (400 cases, 2,000 questions)
nativ-bench user-intent task, 4 options (MASSIVE Hebrew, 2,973 items; the maker's own suite)
