Fastino's open-weight decision model. A DeBERTa-v3-large encoder that scores label sets supplied at runtime, answers single-label, multi-label, yes/no and ordinal questions in one pass, and extracts spans and relations.
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, extract, route
choice, classify, route
Architecture
gliner2
nativ
Fine-tuned from
fastino/gliner2-large-v1
dicta-il/neodictabert
License
apache-2.0
cc-by-4.0
Availability
Open weights + hosted API
Open weights
Hosted by
Fastino
—
Input price
—
—
Decision accuracy
60.2%
96.2%
Calibration error
—
0.010
Valid action rate
—
—
Median latency
38.3 ms
—
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 gliner2-5-decide and nativ-he-decision?
gliner2-5-decide is from Fastino Labs and nativ-he-decision from Yoav Pinto. gliner2-5-decide has open weights and a hosted API; nativ-he-decision has open weights you can download and run. Both answer choice, classify and route questions. Only gliner2-5-decide answers score, noul and extract. gliner2-5-decide is the smaller model, at 340M parameters to 364M. gliner2-5-decide is licensed apache-2.0; nativ-he-decision, cc-by-4.0.
Which is more accurate, gliner2-5-decide or nativ-he-decision?
They report on different suites — gliner2-5-decide 60.2% on Fastino fast-decisions suite (17 datasets), 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, gliner2-5-decide or nativ-he-decision?
gliner2-5-decide: Hosted, price not published, or free to self-host. nativ-he-decision: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run gliner2-5-decide or nativ-he-decision locally?
Yes, both: systemone pull fastino-labs/gliner2-5-decide and systemone pull yoav-pinto/nativ-he-decision download the weights.
Fastino fast-decisions suite (17 datasets)
nativ-bench user-intent task, 4 options (MASSIVE Hebrew, 2,973 items; the maker's own suite)
