Caliber Flex 1B by OnePass. Decision-only model, exactly 1,000,000,000 parameters.
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, noul, score
choice, classify, route
Architecture
llm
nativ
Fine-tuned from
qwen/qwen3.5-0.8b-base
dicta-il/neodictabert
License
caliber-flex-non-commercial-1.0
cc-by-4.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
—
96.2%
Calibration error
—
0.010
Valid action rate
—
—
Median latency
—
—
p95 latency
—
—
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 caliber-flex-1b and nativ-he-decision?
caliber-flex-1b is from onepass and nativ-he-decision from Yoav Pinto. Both have open weights you can download and run. Both answer choice questions. Only caliber-flex-1b answers noul and score. Only nativ-he-decision answers classify and route. caliber-flex-1b is licensed other; nativ-he-decision, cc-by-4.0.
Which is more accurate, caliber-flex-1b or nativ-he-decision?
Only nativ-he-decision publishes an accuracy figure (96.2% on nativ-bench user-intent task, 4 options (MASSIVE Hebrew, 2,973 items; the maker's own suite)); caliber-flex-1b does not, so there is no comparison to make without your own test.
Which is cheaper, caliber-flex-1b or nativ-he-decision?
caliber-flex-1b: Free (open weights). nativ-he-decision: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run caliber-flex-1b or nativ-he-decision locally?
Yes, both: systemone pull onepass/caliber-flex-1b and systemone pull yoav-pinto/nativ-he-decision download the weights.
