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.
A 4B decision model fine-tuned from Qwen3.5-4B-Base. A state and typed questions (choice, yes/no, score) go in; a temperature-calibrated probability for every option comes out of one pass, through a self-hosted /v1/systemone-compatible server. Apache-2.0 weights and code.
Decides
choice, classify, route
choice, score, noul, classify, route
Architecture
nativ
wald
Fine-tuned from
dicta-il/neodictabert
qwen/qwen3.5-4b-base
License
cc-by-4.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
96.2%
88.3%
Calibration error
0.010
0.045
Valid action rate
—
—
Median latency
—
33 ms
p95 latency
—
168 ms
Evaluation suite
nativ-bench user-intent task, 4 options (MASSIVE Hebrew, 2,973 items; the maker's own 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 nativ-he-decision and wald?
nativ-he-decision is from Yoav Pinto and wald from ORG2 AI. Both have open weights you can download and run. Both answer choice, classify and route questions. Only wald answers score and noul. wald reads up to 128K tokens of state, against 1K tokens for nativ-he-decision. nativ-he-decision is the smaller model, at 364M parameters to 4.2B. nativ-he-decision is licensed cc-by-4.0; wald, apache-2.0.
Which is more accurate, nativ-he-decision or wald?
They report on different suites — nativ-he-decision 96.2% on nativ-bench user-intent task, 4 options (MASSIVE Hebrew, 2,973 items; the maker's own suite), wald 88.3% on JevBench public set (231 items; 204 correct), maker's run with the JevBench CLI, v1.2, effort none; items used as a development scoreboard — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, nativ-he-decision or wald?
nativ-he-decision: Free (open weights). wald: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run nativ-he-decision or wald locally?
Yes, both: systemone pull yoav-pinto/nativ-he-decision and systemone pull org2ai/wald download the weights.
JevBench public set (231 items; 204 correct), maker's run with the JevBench CLI, v1.2, effort none; items used as a development scoreboard
Latest version
2026.09
1.2.0
Variants
—
Dockerfile, LICENSE, NOTICE, model-files, reference, server
