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 →★ 25,878
ai-experiments
Measured experiments on decision models. Can an open-weight model re-rank search results? Does a token budget explain a benchmark failure? What does one typed decision cost across model families? Every request and response is recorded, and negative results are published as they came out.
Project facts
- Relationship to Jev
- Research
- Evidence
- Code reference
- Language
- Python
- License
- Not reported; check the repository
- Origin
- Original repository
- Repository status
- Not archived
- Created
- 2026-09-24
- GitHub stars
- 0
- Evidence checked
- 2026-09-27T18:35:42.000Z
- Metadata checked
- 2026-09-27T18:35:42.000Z
- Check status
- current
Stars measure the whole repository, including work unrelated to Jev.
Evidence and scope
Code reference records the linked documentation or source. JevHunt has not independently run or benchmarked this project.
TYPESAFE_API_KEY is set. Both pin a build, because a benchmark that cannot say which build produced a number is not a benchmark. A call that never succeeds is recorded as failed with `score: None`. Nothing is
Evidence commit: e684f9e292c0f05ab30d45431976480b83ede4c4
Discovered through: github-search.
How we review →