A training-free layer that turns an open LLM into a decision model. It reads typed choice, yes/no and score answers from one prefill, removes option-order bias with no labels and, from a few hundred labels, calibrates or fits a closed-form head. Weights stay untouched.
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.
Decides
choice, score, noul, classify, route
choice, score, noul, classify, extract, route
Architecture
anyjev
gliner2
Fine-tuned from
qwen/qwen3-8b
fastino/gliner2-large-v1
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights + hosted API
Hosted by
—
Fastino
Input price
—
—
Decision accuracy
77.1%
60.2%
Calibration error
0.034
—
Valid action rate
—
—
Median latency
—
38.3 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 anyjev and gliner2-5-decide?
anyjev is from Nokia Applied Research and gliner2-5-decide from Fastino Labs. anyjev has open weights you can download and run; gliner2-5-decide has open weights and a hosted API. Both answer choice, score, noul, classify and route questions. Only gliner2-5-decide answers extract. gliner2-5-decide is the smaller model, at 340M parameters to 8.0B.
Which is more accurate, anyjev or gliner2-5-decide?
They report on different suites — anyjev 77.1% on LocalLLaMA/typed-decisions (2,000 held-out decisions), L2 with 300 labels per question, gliner2-5-decide 60.2% on Fastino fast-decisions suite (17 datasets) — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, anyjev or gliner2-5-decide?
anyjev: Free (open weights). gliner2-5-decide: Hosted, price not published, or free to self-host. Open weights cost nothing per call beyond your own hardware.
Can I run anyjev or gliner2-5-decide locally?
Yes, both: systemone pull nokia-applied-research/anyjev and systemone pull fastino-labs/gliner2-5-decide download the weights.
Evaluation suite
LocalLLaMA/typed-decisions (2,000 held-out decisions), L2 with 300 labels per question
