HEAD TO HEAD
Kotoba Labs: open-jev-deberta-v3-large and Rizzo AI Academy: rizzo-flow, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | kotoba-labs/open-jev-deberta-v3-large | rizzo-ai-academy/rizzo-flow |
|---|---|---|
| Summary | 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. | 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, score, noul, classify | choice, score, noul, classify, route |
| Architecture | open-jev | rizzo-flow |
| Fine-tuned from | microsoft/deberta-v3-large | xhtoken/spark-x2.5-4b |
| License | apache-2.0 | apache-2.0 |
| Availability | Open weights | Open weights |
| Hosted by | — | — |
| Input price | — | — |
| Decision accuracy | 85.4% | 64.8% |
| Calibration error | 0.022 | 0.112 |
| Valid action rate | — | — |
| Median latency | 28 ms | 195 ms |
| p95 latency | — | — |
| Evaluation suite | Kotoba held-out test, seen question types (banking77, SST-5, BoolQ) | LocalLLaMA/typed-decisions test split (400 cases, 2,000 decisions), Q8_0 |
| Latest version | 2026.09 | 2026.09 |
| Variants | — | lora, LICENSE |
| Size of latest version | 1.6 GB | 22.0 GB |
| Files | 9 | 17 |
| Downloads | 2 | 2 |
| Stars | 0 | 0 |
| Tags | system-one, encoder, deberta, public-data, 434m | system-one, spark, lora, gguf, llama-cpp, 4b |
| Updated | Oct 7, 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.
open-jev-deberta-v3-large is from Kotoba Labs and rizzo-flow from Rizzo AI Academy. Both have open weights you can download and run. Both answer choice, score, noul and classify questions. Only rizzo-flow answers route. open-jev-deberta-v3-large is the smaller model, at 434M parameters to 4.0B.
They report on different suites — open-jev-deberta-v3-large 85.4% on Kotoba held-out test, seen question types (banking77, SST-5, BoolQ), 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.
open-jev-deberta-v3-large: Free (open weights). rizzo-flow: Free (open weights). Open weights cost nothing per call beyond your own hardware.
By their publishers’ figures, open-jev-deberta-v3-large answers in about 28 ms at the median and rizzo-flow in about 195 ms — measured on different hardware, so treat it as a rough guide.
Yes, both: systemone pull kotoba-labs/open-jev-deberta-v3-large and systemone pull rizzo-ai-academy/rizzo-flow download the weights.