An open family of System One models. A LoRA adapter plus a pointer head on a frozen Qwen base returns a distribution per typed question in one forward pass, serves TypeSafe's /v1/systemone contract, and ships a fitted temperature with every checkpoint.
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
kev
nativ
Fine-tuned from
qwen/qwen3.5-4b-base
dicta-il/neodictabert
License
apache-2.0
cc-by-4.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
83.8%
96.2%
Calibration error
0.042
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 kev and nativ-he-decision?
kev is from Jared Palmer and nativ-he-decision from Yoav Pinto. Both have open weights you can download and run. Both answer choice, classify and route questions. Only kev answers score and noul. kev reads up to 8K tokens of state, against 1K tokens for nativ-he-decision. nativ-he-decision is the smaller model, at 364M parameters to 4.0B. kev is licensed apache-2.0; nativ-he-decision, cc-by-4.0.
Which is more accurate, kev or nativ-he-decision?
They report on different suites — kev 83.8% on transfer-v4 (locked, out of domain), 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, kev or nativ-he-decision?
kev: Free (open weights). nativ-he-decision: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run kev or nativ-he-decision locally?
Yes, both: systemone pull jared-palmer/kev and systemone pull yoav-pinto/nativ-he-decision download the weights.
transfer-v4 (locked, out of domain)
nativ-bench user-intent task, 4 options (MASSIVE Hebrew, 2,973 items; the maker's own suite)
