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 local, Jev-compatible decision model from Rizzo AI Academy. XHToken's Spark-X2.5-4B with a merged typed-decisions LoRA, run on llama.cpp; yes/no, choice and score questions share one prefill of the state and are read from the answer-letter logits. No text is generated.
Decides
choice, classify, route
choice, score, noul, classify, route
Architecture
nativ
rizzo-flow
Fine-tuned from
dicta-il/neodictabert
xhtoken/spark-x2.5-4b
License
cc-by-4.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
96.2%
64.8%
Calibration error
0.010
0.112
Valid action rate
—
—
Median latency
—
195 ms
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 nativ-he-decision and rizzo-flow?
nativ-he-decision is from Yoav Pinto and rizzo-flow from Rizzo AI Academy. Both have open weights you can download and run. Both answer choice, classify and route questions. Only rizzo-flow answers score and noul. nativ-he-decision is the smaller model, at 364M parameters to 4.0B. nativ-he-decision is licensed cc-by-4.0; rizzo-flow, apache-2.0.
Which is more accurate, nativ-he-decision or rizzo-flow?
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), rizzo-flow 64.8% on LocalLLaMA/typed-decisions test split (400 cases, 2,000 decisions), Q8_0 — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, nativ-he-decision or rizzo-flow?
nativ-he-decision: Free (open weights). rizzo-flow: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run nativ-he-decision or rizzo-flow locally?
Yes, both: systemone pull yoav-pinto/nativ-he-decision and systemone pull rizzo-ai-academy/rizzo-flow download the weights.
Evaluation suite
nativ-bench user-intent task, 4 options (MASSIVE Hebrew, 2,973 items; the maker's own suite)
LocalLLaMA/typed-decisions test split (400 cases, 2,000 decisions), Q8_0
