Ruoxi Qiu's decision adapter for a frozen Qwen3.8-27B: a 6.3M-parameter LoRA on the last four text layers plus a numerical prior. One shared state feeds isolated question branches; each Choice, Noul or Score question returns probabilities with no generated tokens.
French medical decision models by Bofeng Huang: pass a patient message or clinical note, a question and candidate answers, and get one probability per answer from a single forward pass. Choice, score and noul. A research model, not a medical device.
Decides
choice, noul, score
choice, score, noul
Architecture
canopy-jev
docto-decision
Fine-tuned from
qwen/qwen3.8-27b
qwen/qwen3.5-4b
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
87.4%
88.9%
Calibration error
—
0.079
Valid action rate
—
—
Median latency
—
47 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 canopy-jev and docto-decision?
canopy-jev is from Ruoxi Qiu and docto-decision from Bofeng Huang. Both have open weights you can download and run. Both answer choice, noul and score questions. docto-decision is the smaller model, at 4.7B parameters to 27B.
Which is more accurate, canopy-jev or docto-decision?
They report on different suites — canopy-jev 87.4% on JevBench public set (231 items; 202 correct), the maker's own fresh zero-shot run with the upstream scorer, docto-decision 88.9% on Docto Decision Bench fr v0.1 (12 tasks, mostly silver labels; the maker's own benchmark) — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, canopy-jev or docto-decision?
canopy-jev: Free (open weights). docto-decision: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run canopy-jev or docto-decision locally?
Yes, both: systemone pull camellia86/canopy-jev and systemone pull bofeng-huang/docto-decision download the weights.
Evaluation suite
JevBench public set (231 items; 202 correct), the maker's own fresh zero-shot run with the upstream scorer
Docto Decision Bench fr v0.1 (12 tasks, mostly silver labels; the maker's own benchmark)
