Loading comparison…
COMPARE
What each one decides, how well calibrated it is, how fast it answers and what it costs to pull. Up to 4 at a time; the better value in each row is marked.
| Property | cua/cua-s1-forms | alibi-serikbay/jevk5 |
|---|---|---|
| Summary | 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. | Qwen3.5-4B with a merged rank-16 LoRA distilled from 17,408 teacher questions and 30,000 public training rows, read out as a temperature-scaled softmax over answer-letter logits. Also in 9B, 2B and GGUF. |
| Decides | choice | choice, score, noul, classify, route |
| Architecture | cua-s1 | jevk5 |
| Fine-tuned from | — | qwen/qwen3.5-4b |
| License | mit | apache-2.0 |
| Availability | Open weights | Open weights |
| Hosted by | — | — |
| Input price | — | — |
| Decision accuracy | 100.0% | 78.4% |
| Calibration error | — | 0.035 |
| Valid action rate | — | — |
| Median latency | — | 13.2 ms |
| p95 latency | — | — |
| Evaluation suite | 196-decision evaluation on real forms | JevBench public hard tier |
| Latest version | 2026.09 | 0.3.0 |
| Variants | — | SHA256SUMS |
| Size of latest version | 5.4 MB | 7.9 GB |
| Files | 3 | 8 |
| Downloads | 0 | 0 |
| Stars | 0 | 0 |
| Tags | system-one, computer-use, forms, tiny, 706k | system-one, qwen, lora, distilled, 4b |
| Updated | Sep 26, 2026 | Sep 26, 2026 |