Berget AI's System One model for gating agent commands: a LoRA adapter and a fine-tuned joint schema head on Cloudflare's Clef-Flash that answer noul, choice and score questions over a state in one forward pass. Trained on Swedish and English operations decisions.
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
noul, choice, score
choice, classify, route
Architecture
clef
nativ
Fine-tuned from
cloudflare/clef-flash
dicta-il/neodictabert
License
apache-2.0
cc-by-4.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
97.0%
96.2%
Calibration error
—
0.010
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 bev and nativ-he-decision?
bev is from Berget AI and nativ-he-decision from Yoav Pinto. Both have open weights you can download and run. Both answer choice questions. Only bev answers noul and score. Only nativ-he-decision answers classify and route. nativ-he-decision is the smaller model, at 364M parameters to 9.0B. bev is licensed apache-2.0; nativ-he-decision, cc-by-4.0.
Which is more accurate, bev or nativ-he-decision?
They report on different suites — bev 97.0% on Berget held-out risk split (16,902 questions; same operations-traffic corpora as training), 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, bev or nativ-he-decision?
bev: Free (open weights). nativ-he-decision: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run bev or nativ-he-decision locally?
Yes, both: systemone pull berget-ai/bev and systemone pull yoav-pinto/nativ-he-decision download the weights.
Evaluation suite
Berget held-out risk split (16,902 questions; same operations-traffic corpora as training)
nativ-bench user-intent task, 4 options (MASSIVE Hebrew, 2,973 items; the maker's own suite)
