A 0.6B decision model on Qwen3-0.6B-Base from the DocsGPT team. Give it a state and typed questions (yes/no, a choice of up to 16 options, a 3- or 4-level score) and it returns calibrated probabilities in one pass, for RAG and agent checks.
SimpleJev's LoRA adapters and pointer head that turn an open model into a typed-decision model: given a state, a question and the options, it returns a probability for every option without generating text. This entry is the Qwen3.8-27B release; the others are linked.
Decides
choice, score, noul, classify, rank, route
choice, score, noul
Architecture
arc-decide
jevany
Fine-tuned from
qwen/qwen3-0.6b-base
qwen/qwen3.8-27b
License
mit
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
—
86.0%
Calibration error
—
0.026
Valid action rate
—
—
Median latency
—
—
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 arc-decide and jevany?
arc-decide is from Arc53 and jevany from SimpleJev. Both have open weights you can download and run. Both answer choice, score and noul questions. Only arc-decide answers classify, rank and route. arc-decide is the smaller model, at 600M parameters to 27B. arc-decide is licensed mit; jevany, apache-2.0.
Which is more accurate, arc-decide or jevany?
Only jevany publishes an accuracy figure (86.0% on SimpleJev Transfer suite (1,046 decisions from seven cross-domain datasets plus robustness slices; the maker's own)); arc-decide does not, so there is no comparison to make without your own test.
Which is cheaper, arc-decide or jevany?
arc-decide: Free (open weights). jevany: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run arc-decide or jevany locally?
Yes, both: systemone pull arc53/arc-decide and systemone pull simplejev/jevany download the weights.
Evaluation suite
—
SimpleJev Transfer suite (1,046 decisions from seven cross-domain datasets plus robustness slices; the maker's own)
