A 0.4B decision model made from the first 20 layers of Qwen3-0.6B plus a small attention head that compares options. It answers choice, noul and score questions in one pass, and choice order cannot change its answer by construction. Non-commercial licence.
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.
Decides
choice, score, noul
choice, noul, score
Architecture
bev-decider
canopy-jev
Fine-tuned from
qwen/qwen3-0.6b
qwen/qwen3.8-27b
License
cc-by-nc-4.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
74.7%
87.4%
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 bev-decider and canopy-jev?
bev-decider is from Avishek Biswas and canopy-jev from Ruoxi Qiu. Both have open weights you can download and run. Both answer choice, score and noul questions. bev-decider is the smaller model, at 478M parameters to 27B. bev-decider is licensed cc-by-nc-4.0; canopy-jev, apache-2.0.
Which is more accurate, bev-decider or canopy-jev?
They report on different suites — bev-decider 74.7% on avbiswas/bev-decision test split (5,000 held-out questions over 2,617 states), the maker's own, canopy-jev 87.4% on JevBench public set (231 items; 202 correct), the maker's own fresh zero-shot run with the upstream scorer — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, bev-decider or canopy-jev?
bev-decider: Free (open weights). canopy-jev: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run bev-decider or canopy-jev locally?
Yes, both: systemone pull avishek-biswas/bev-decider and systemone pull camellia86/canopy-jev download the weights.
—
Evaluation suite
avbiswas/bev-decision test split (5,000 held-out questions over 2,617 states), the maker's own
JevBench public set (231 items; 202 correct), the maker's own fresh zero-shot run with the upstream scorer
