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.
A 149.6M ModernBERT-base cross-encoder with a scalar head that picks an agent's next action or tool from labels and descriptions you supply. Each option is scored against the state and a softmax ranks the set; no text is generated. Experimental preview, not calibrated.
Decides
choice, noul, score
choice, rank, route
Architecture
canopy-jev
modernjev
Fine-tuned from
qwen/qwen3.8-27b
answerdotai/modernbert-base
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
87.4%
73.4%
Calibration error
—
—
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 canopy-jev and modernjev-decide?
canopy-jev is from Ruoxi Qiu and modernjev-decide from Maziyar Panahi. Both have open weights you can download and run. Both answer choice questions. Only canopy-jev answers noul and score. Only modernjev-decide answers rank and route. modernjev-decide is the smaller model, at 150M parameters to 27B.
Which is more accurate, canopy-jev or modernjev-decide?
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, modernjev-decide 73.4% on AgentToolDecisions-180K held-out test, next action type (1,158 decisions, 3 choices; the maker's split) — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, canopy-jev or modernjev-decide?
canopy-jev: Free (open weights). modernjev-decide: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run canopy-jev or modernjev-decide locally?
Yes, both: systemone pull camellia86/canopy-jev and systemone pull maziyar-panahi/modernjev-decide download the weights.
—
Evaluation suite
JevBench public set (231 items; 202 correct), the maker's own fresh zero-shot run with the upstream scorer
AgentToolDecisions-180K held-out test, next action type (1,158 decisions, 3 choices; the maker's split)
