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 local, Jev-compatible decision model from Rizzo AI Academy. XHToken's Spark-X2.5-4B with a merged typed-decisions LoRA, run on llama.cpp; yes/no, choice and score questions share one prefill of the state and are read from the answer-letter logits. No text is generated.
Decides
choice
choice, score, noul, classify, route
Architecture
cua-s1
rizzo-flow
Fine-tuned from
—
xhtoken/spark-x2.5-4b
License
mit
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
100.0%
64.8%
Calibration error
—
0.112
Valid action rate
—
—
Median latency
—
195 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 rizzo-flow?
cua-s1-forms is from Cua and rizzo-flow from Rizzo AI Academy. Both have open weights you can download and run. Both answer choice questions. Only rizzo-flow answers score, noul, classify and route. cua-s1-forms is the smaller model, at 706K parameters to 4.0B. cua-s1-forms is licensed mit; rizzo-flow, apache-2.0.
Which is more accurate, cua-s1-forms or rizzo-flow?
They report on different suites — cua-s1-forms 100.0% on 196-decision evaluation on real forms, rizzo-flow 64.8% on LocalLLaMA/typed-decisions test split (400 cases, 2,000 decisions), Q8_0 — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, cua-s1-forms or rizzo-flow?
cua-s1-forms: Free (open weights). rizzo-flow: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run cua-s1-forms or rizzo-flow locally?
Yes, both: systemone pull cua/cua-s1-forms and systemone pull rizzo-ai-academy/rizzo-flow download the weights.
196-decision evaluation on real forms
LocalLLaMA/typed-decisions test split (400 cases, 2,000 decisions), Q8_0
