Trained models

Own decision heads

Downloadable weights for local serving and research.

Inference engines

Reuse base weights

Read logits or broadcast KV cache to produce Jev-like outputs.

Compatibility

Match the API shape

The interface can match while the underlying quality differs.

Projects to start with

ProjectTypeBase / feature
Mapika/decider-2btrained modelQwen3.5-2B, Choice / Score / Noul
Layatrained modelModernBERT, 421M
NanoJevresearchQwen3-0.6B, games and mazes
zhihz/openjevlocal previewQwen3-4B, bilingual experiments
SemIfengineNative candidate-probability readout

Which path should you pick?

  • General local API: start with decider-2b.
  • Mac experiments: look at SemIf, open-jev and jev-on-a-laptop.
  • Bilingual research: test zhihz/openjev with your own data.
  • Teaching demos: look at NanoJev and jevlike.
Shared disclaimer: These projects are not official TypeSafe AI Jev weights. Speed, accuracy and comparisons should be treated as author-reported until independently reproduced.

Model cards

Each model page records weights, base model, license, context, language, question types, calibration evidence and failure cases.