Jev by @typesafeai is now on OpenRouter, in beta. Jev is a System One model. Instead of generating text, it takes your app's state plus a typed question and returns a typed decision with a probability attached. There is no JSON prompting, parsing layer, and nothing to validate Show more
Cookbook: Double-Checking Citations
Catch wrong and hallucinated citations against RFC 7519 with a string match plus one Choice question: four accurate citations verified at 0.93 or more and all four planted failures caught.
- Category
- Cookbooks & Demos
- Published by
- TypeSafe AI
- Author
- —
- Added
- 2026-09-20
Tagsofficialcookbookcitationsguardrailsconfidence
Highlights
- Eight citations across RFC 7519's 45 sections: four verified at confidence 0.93+, one fabricated, one contradicted, two unsupported.
- An exact string match catches missing quotes with no model call, while the contradicted example still scored 0.99.
- Unsupported citations at 0.27 and 0.56 fell below the 0.8 auto-accept gate and went to a human.
Watch out
The string match is exact after normalization, so truncated or lightly reworded quotes come back as fabricated.
More like this
Score retrieved passages with four Noul questions each, then route in code: prompt injections dropped, false premises kept as conflicts, and evidence assembled separately for the generator.
Cookbooks & Demos#official#cookbook#search
Official
Semantic search over GitHub's Terms of Service: one request ranks all 218 lines with a Choice while a Noul checks whether the document contains an answer at all, including when it should say no.
Cookbooks & Demos#official#cookbook#search
Official
A two-stage extraction cascade: a mini model extracts, a Noul battery verifies each field in one request, and a 0.7 gate escalates to a reasoning model, sitting on the cost/quality frontier.
Cookbooks & Demos#official#cookbook#extraction
Official
From the community
Posts from builders shipping with Jev right now.
Jev lands on OpenRouter
700 leads scored for $0.09
JEV is INSANE. We gave it 700 high-intent leads and personalised outreach messages. In 40 seconds, it predicted how each message would perform, assigned a confidence score and detected lead-message mismatches. All for just $0.09. JEV can also score leads, analyse buying Show more