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.
FrontiersMind's decision model on a base it pre-trained from scratch. Each question reads the state and its own tokens once and a 256-dimension pointer head scores the options; Choice and Score take up to 255 entries, and nothing is generated.
Decides
choice
choice, score, noul
Architecture
cua-s1
lumma-fev
Fine-tuned from
—
frontiersmind/lumma-0.6b-base
License
mit
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
100.0%
64.0%
Calibration error
—
—
Valid action rate
—
—
Median latency
—
45.8 ms
p95 latency
—
—
Evaluation suite
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 lumma-fev?
cua-s1-forms is from Cua and lumma-fev from FrontiersMind. Both have open weights you can download and run. Both answer choice questions. Only lumma-fev answers score and noul. cua-s1-forms is the smaller model, at 706K parameters to 649M. cua-s1-forms is licensed mit; lumma-fev, apache-2.0.
Which is more accurate, cua-s1-forms or lumma-fev?
They report on different suites — cua-s1-forms 100.0% on 196-decision evaluation on real forms, lumma-fev 64.0% on typed-decisions (maker's table; split not stated) — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, cua-s1-forms or lumma-fev?
cua-s1-forms: Free (open weights). lumma-fev: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run cua-s1-forms or lumma-fev locally?
Yes, both: systemone pull cua/cua-s1-forms and systemone pull frontiersmind/lumma-fev download the weights.
