A 2.8 MB byte-level transformer for GUI form filling. For each form element it returns one probability per typed option (fill an entity, check, click or skip), using jevlike's option-attention head.
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.
Decides
choice
choice, score, noul, classify
Architecture
cua-s1
hopper
Fine-tuned from
—
qwen/qwen3.5-4b
License
mit
Research and demo use only (training data includes RACE, non-commercial); serving code Apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
100.0%
68.5%
Calibration error
—
0.102
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 cua-s1-forms and hopper?
cua-s1-forms is from Cua and hopper from HopitAI. Both have open weights you can download and run. Both answer choice questions. Only hopper answers score, noul and classify. cua-s1-forms is the smaller model, at 706K parameters to 4.0B. cua-s1-forms is licensed mit; hopper, other.
Which is more accurate, cua-s1-forms or hopper?
They report on different suites — cua-s1-forms 100.0% on 196-decision evaluation on real forms, hopper 68.5% on JevBench public hard tier (111 items, measured by the authors) — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, cua-s1-forms or hopper?
cua-s1-forms: Free (open weights). hopper: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run cua-s1-forms or hopper locally?
Yes, both: systemone pull cua/cua-s1-forms and systemone pull hopit-ai/hopper download the weights.
Evaluation suite
196-decision evaluation on real forms
JevBench public hard tier (111 items, measured by the authors)
