Local alternative
Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.
Details →★ 24,501
open-medical-jev
Jev/OpenJev-level results from frozen open models — within ~2 pts of Jev on national medical exams (level with the OpenJev run; ≈3× Laya's released checkpoints); no fine-tuning, no distillation, no corpus — so newer open models swap in without retraining. Code + recipe only.
Project facts
- Relationship to Jev
- Research
- Evidence
- Documented
- Language
- Not reported
- License
- MIT
- Origin
- Original repository
- Repository status
- Not archived
- Created
- 2026-09-25
- GitHub stars
- 1
- Evidence checked
- 2026-09-26T10:34:38.000Z
- Metadata checked
- 2026-09-26T10:34:38.000Z
- Check status
- current
Stars measure the whole repository, including work unrelated to Jev.
Evidence and scope
Documented records the linked documentation or source. JevHunt has not independently run or benchmarked this project.
| Item set | 27B | 35B | 2-model mean fusion | router: unanimous coverage @ acc | Jev (hosted) |
Evidence commit: f6d4f79aa2cb5566168c3eb837939d20e92d4657
Discovered through: typesafe-field-guide.
How we review →