Docs
OfficialJev Practices & Patterns
Noul
The yes/no primitive: one probability from 0 to 1, where the value is the answer and no separate confidence is needed. True and false criteria pin down subtle boundaries.
- Format
- Docs
- Reading time
- 2 min
- Published by
- TypeSafe AI
- Last verified
- Sep 2026
Works forJev for Decision gating
Used inJev for Software & DevTools
Highlights
- Recorded escalation answers ranged from 0.02 for a resolved complaint to 0.99 for a demand to talk to a person.
- Optional criteria.true and criteria.false descriptions resolve boundary cases.
- Threshold in code: 0.5 when either error is equally easy to act on, higher when a false yes is expensive.
- Ask one question per Noul; middle values are a natural hand-off to a person.
Quickstart
python
wants_human = response.answers["is_human_escalation"].noul > 0.9
if wants_human:
route_to_agent(ticket)
else:
route_to_bot(ticket)Watch out
A Noul value is the probability of yes, not a degree scale, so it does not measure how much or how strongly.
Questions about this entry
Short answers drawn from the linked project or page.
When a false yes is expensive, set the threshold higher. The page advises 0.5 when either error is equally easy to act on, and a higher bar when acting wrongly costs more than waiting.
Yes, but ask one question per Noul. Middle values are a natural hand-off to a person, and the recorded escalation example ranged from 0.02 for a resolved complaint to 0.99 for a demand to talk to a person.
Type is always noul, instructions is the yes/no question or statement to judge, and criteria is an optional object with true and false descriptions of what a yes and a no mean.
No. A Noul's probability distribution has only two outcomes, so the single noul value describes it completely, unlike a Choice or Score which spreads probability over several options.
That the model gives yes and no similar probability, the kind of message where a decision needs a threshold in your code rather than a clear answer.
No. You choose ids such as is_human_escalation and is_repeat_contact, and they are not sent to the model; each answer comes back under the same id.
Yes. Instructions can be a string, an object, or an array; use an object when the question needs data alongside it, such as a record to compare the state against.
More Jev resources like this
Pick one option from a set you define. Returns the option, the full probability distribution, and 0-1 confidence, with up to 255 options and parallel questions that barely add latency.
Jev Practices & PatternsDocs#official#primitives#choice
Official
Use the answer to decide what to do and confidence to decide whether to act: the voice-banking example routes below 0.6 to a human and needs 0.85 or more to auto-approve a transfer.
Jev Practices & PatternsDocs#official#patterns#confidence
Official
Classify intent with a Choice and complexity with a Score, then send each branch to deterministic code, a specialist model, or a person, with a 0.5 intent-confidence floor.
Jev Practices & PatternsDocs#official#patterns#routing
Official
A maintained list of jev-1.13's known failure modes, literal reading, unreliable counting, dates as text, indirection, context rot, and contradictory criteria, each with a guardrail.
Jev Practices & PatternsDocs#official#docs#evaluation
Official
Rate content against 2 to 10 ordered levels. Returns a probability-weighted score that can land between levels, per-level probabilities, and confidence, with the arithmetic left to code.
Jev Practices & PatternsDocs#official#primitives#score
Official
Ask everything the system might need in one request, then let code throw away what it does not use. The ticket-triage example sends a category plus four speculative questions in parallel.
Jev Practices & PatternsDocs#official#patterns#batching
Official
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.