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.
Unofficial NoulXP package of Convai Innovations' Laya typed-decisions (Apache-2.0), fp32-o23: checked against the model's own answers, none changed. Not made by Convai Innovations.
Decides
choice, noul, score
choice, score, noul, classify, route
Architecture
canopy-jev
laya
Fine-tuned from
qwen/qwen3.8-27b
convai-innovations/laya
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
87.4%
—
Calibration error
—
—
Valid action rate
—
—
Median latency
—
126 ms
p95 latency
—
2185 ms
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 laya-typed-decisions-noulxp?
canopy-jev is from Ruoxi Qiu and laya-typed-decisions-noulxp from Convai Innovations. Both have open weights you can download and run. Both answer choice, noul and score questions. Only laya-typed-decisions-noulxp answers classify and route. laya-typed-decisions-noulxp is the smaller model, at 421M parameters to 27B.
Which is more accurate, canopy-jev or laya-typed-decisions-noulxp?
Only canopy-jev publishes an accuracy figure (87.4% on JevBench public set (231 items; 202 correct), the maker's own fresh zero-shot run with the upstream scorer); laya-typed-decisions-noulxp does not, so there is no comparison to make without your own test.
Which is cheaper, canopy-jev or laya-typed-decisions-noulxp?
canopy-jev: Free (open weights). laya-typed-decisions-noulxp: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run canopy-jev or laya-typed-decisions-noulxp locally?
Yes, both: systemone pull camellia86/canopy-jev and systemone pull convai-innovations/laya-typed-decisions-noulxp download the weights.
Evaluation suite
JevBench public set (231 items; 202 correct), the maker's own fresh zero-shot run with the upstream scorer
Measured by System One Models on 2026-10-05: one question per request, 4 threads of a Xeon Gold 6342
