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 | iapp-technology/openthai-systemone | wfzyx/von |
|---|---|---|
| Summary | A Thai-and-English System One model. Qwen3.5-0.8B continued-pretrained on about 5B Thai tokens, its language-model head replaced by a 256-way slot head (slot 255 abstains), trained on 2–3M decision examples and temperature-calibrated per question type. | A ModernBERT-large encoder with an option-marker head. Premise and options are packed into one sequence and each option's marker is scored in a single bidirectional pass; version 1.2 makes the scoring order-invariant. |
| Decides | choice, score, noul, classify, route | choice, score, noul, classify |
| Architecture | openthai | von |
| Fine-tuned from | qwen/qwen3.5-0.8b-base | answerdotai/modernbert-large |
| License | apache-2.0 | apache-2.0 |
| Availability | Open weights | Open weights |
| Hosted by | — | — |
| Input price | — | — |
| Decision accuracy | 74.3% | 63.9% |
| Calibration error | — | 0.045 |
| Valid action rate | — | — |
| Median latency | 40 ms | 18 ms |
| p95 latency | — | — |
| Evaluation suite | Bespoke public subsets (13, macro average) | JevBench public standard tier |
| Latest version | 0.3.0 | 1.2.0 |
| Variants | — | — |
| Size of latest version | 1.4 GB | 2.9 GB |
| Files | 2 | 2 |
| Downloads | 0 | 0 |
| Stars | 0 | 0 |
| Tags | system-one, thai, multilingual, qwen, 0.8b | system-one, encoder, modernbert, 395m |
| Updated | Sep 26, 2026 | Sep 26, 2026 |