A rank-16 LoRA on Qwen3.5-4B built for the JevBench setting. A document, a policy and a question go in; a calibrated distribution over the option letters comes out of one forward pass. Research and demo use only.
An independent Jev-shaped reproduction on DeBERTa-v3-large. One state and any number of choice, score and yes/no questions go in, and a calibrated distribution per question comes out of one forward pass. Public gold labels only.
Decides
choice, score, noul, classify
choice, score, noul, classify
Architecture
hopper
open-jev
Fine-tuned from
qwen/qwen3.5-4b
microsoft/deberta-v3-large
License
Research and demo use only (training data includes RACE, non-commercial); serving code Apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
68.5%
85.4%
Calibration error
0.102
0.022
Valid action rate
—
—
Median latency
—
28 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 hopper and open-jev-deberta-v3-large?
hopper is from HopitAI and open-jev-deberta-v3-large from Kotoba Labs. Both have open weights you can download and run. Both answer choice, score, noul and classify questions. open-jev-deberta-v3-large is the smaller model, at 434M parameters to 4.0B. hopper is licensed other; open-jev-deberta-v3-large, apache-2.0.
Which is more accurate, hopper or open-jev-deberta-v3-large?
They report on different suites — hopper 68.5% on JevBench public hard tier (111 items, measured by the authors), open-jev-deberta-v3-large 85.4% on Kotoba held-out test, seen question types (banking77, SST-5, BoolQ) — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, hopper or open-jev-deberta-v3-large?
hopper: Free (open weights). open-jev-deberta-v3-large: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run hopper or open-jev-deberta-v3-large locally?
Yes, both: systemone pull hopit-ai/hopper and systemone pull kotoba-labs/open-jev-deberta-v3-large download the weights.
Evaluation suite
JevBench public hard tier (111 items, measured by the authors)
Kotoba held-out test, seen question types (banking77, SST-5, BoolQ)
