Loading comparison…
HEAD TO HEAD
Nokia: anyjev and Yoav Pinto: nativ-he-decision, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | nokia/anyjev | yoav-pinto/nativ-he-decision |
|---|---|---|
| Summary | A training-free layer that turns an open LLM into a decision model. It reads typed choice, yes/no and score answers from one prefill, removes option-order bias with no labels and, from a few hundred labels, calibrates or fits a closed-form head. Weights stay untouched. | A 364M Hebrew decision encoder on NeoDictaBERT: give it a state, a question and 2 to 7 free-text options and it returns a probability for each option in one forward pass, about 70 ms on a laptop CPU. Hebrew only. |
| Decides | choice, score, noul, classify, route | choice, classify, route |
| Architecture | anyjev | nativ |
| Fine-tuned from | qwen/qwen3-8b | dicta-il/neodictabert |
| License | apache-2.0 | cc-by-4.0 |
| Availability | Open weights | Open weights |
| Hosted by | — | — |
| Input price | — | — |
| Decision accuracy | 77.1% | 96.2% |
| Calibration error | 0.034 | 0.010 |
| Valid action rate | — | — |
| Median latency | — | — |
| p95 latency | — | — |
| Evaluation suite | LocalLLaMA/typed-decisions (2,000 held-out decisions), L2 with 300 labels per question | nativ-bench user-intent task, 4 options (MASSIVE Hebrew, 2,973 items; the maker's own suite) |
| Latest version | 0.2.0 | 2026.09 |
| Variants | — | — |
| Size of latest version | — | 697.3 MB |
| Files | 0 | 5 |
| Downloads | 0 | 0 |
| Stars | 1 | 0 |
| Tags | system-one, qwen, training-free, calibrated, vllm, 8b | system-one, hebrew, encoder, cpu, 364m |
| Updated | Oct 25, 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.
anyjev is from Nokia and nativ-he-decision from Yoav Pinto. Both have open weights you can download and run. Both answer choice, classify and route questions. Only anyjev answers score and noul. nativ-he-decision is the smaller model, at 364M parameters to 8.0B. anyjev is licensed apache-2.0; nativ-he-decision, cc-by-4.0.
They report on different suites — anyjev 77.1% on LocalLLaMA/typed-decisions (2,000 held-out decisions), L2 with 300 labels per question, nativ-he-decision 96.2% on nativ-bench user-intent task, 4 options (MASSIVE Hebrew, 2,973 items; the maker's own suite) — so the numbers do not rank them. Test both on your own labelled examples.
anyjev: Free (open weights). nativ-he-decision: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Yes, both: systemone pull nokia/anyjev and systemone pull yoav-pinto/nativ-he-decision download the weights.