HEAD TO HEAD
Mapika: decider and Kotoba Labs: open-jev-deberta-v3-large, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | mapika/decider | kotoba-labs/open-jev-deberta-v3-large |
|---|---|---|
| Summary | A family of open System One models by Mark Marosi that softmax option-letter logits at an answer slot: decider-0.8b, 2b, 4b and 35b-a3b on Qwen3.5 bases, decider-12b on Gemma-4-12B-it, and training-free readouts of larger chat models. This page carries decider-2b v11. | An independent Jev-shaped reproduction on DeBERTa-v3-large. One state and any number of choice, score and yes/no questions go in, and a calibrated distribution per question comes out of one forward pass. Public gold labels only. |
| Decides | choice, score, noul, classify, route | choice, score, noul, classify |
| Architecture | decider | open-jev |
| Fine-tuned from | qwen/qwen3.5-2b-base | microsoft/deberta-v3-large |
| License | apache-2.0 | apache-2.0 |
| Availability | Open weights | Open weights |
| Hosted by | — | — |
| Input price | — | — |
| Decision accuracy | 75.2% | 85.4% |
| Calibration error | — | 0.022 |
| Valid action rate | — | — |
| Median latency | — | 28 ms |
| p95 latency | — | — |
| Evaluation suite | Decider regression set, 28 held-out tasks (decider-2b v11) | Kotoba held-out test, seen question types (banking77, SST-5, BoolQ) |
| Latest version | v11 | 2026.09 |
| Variants | — | — |
| Size of latest version | 5.4 GB | 1.6 GB |
| Files | 6 | 9 |
| Downloads | 66 | 2 |
| Stars | 0 | 0 |
| Tags | system-one, qwen, gemma, rl, calibrated, 1.9b | system-one, encoder, deberta, public-data, 434m |
| Updated | Nov 15, 2026 | Oct 7, 2026 |
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.
decider is from Mapika and open-jev-deberta-v3-large from Kotoba Labs. Both have open weights you can download and run. Both answer choice, score, noul and classify questions. Only decider answers route. decider reads up to 32K tokens of state, against 512 tokens for open-jev-deberta-v3-large. open-jev-deberta-v3-large is the smaller model, at 434M parameters to 1.9B.
They report on different suites — decider 75.2% on Decider regression set, 28 held-out tasks (decider-2b v11), open-jev-deberta-v3-large 85.4% on Kotoba held-out test, seen question types (banking77, SST-5, BoolQ) — so the numbers do not rank them. Test both on your own labelled examples.
decider: Free (open weights). open-jev-deberta-v3-large: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Yes, both: systemone pull mapika/decider and systemone pull kotoba-labs/open-jev-deberta-v3-large download the weights.