Unverified50% confidenceFactExact time
Instead of generating text for routing and classification, Jev directly returns structured decisions: a choice from a fixed set of options, a score on an ordered scale, or a probability for a yes/no statement.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
Related Entities
Related Claims
UnverifiedJev only generates yes/no answers, answers chosen from a list, or scores on a scale, with each answer accompanied by a probability, and outputs no free-form text.73% similarUnverifiedEach time Jev returns a decision, it attaches a confidence value, expressed as a number indicating how certain the model is in its judgment among the options.73% similarUnverifiedThe Jev model only makes three types of judgments: selecting one from options, giving a score, or outputting yes/no, and each judgment comes with a confidence level.71% similarUnverifiedJev allows the definition of three output object types: Choice (a probability distribution over unordered fixed options), Score (an ordered sliding scale), and Null (a yes/no type on a 0-to-1 scale).71% similarVerifiedJev is a type of front-end model called a 'System 1 model' — fast, cheap, and structured, designed specifically for decision-making; it does not generate text but instead returns probability scores within a predefined set of options.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1000505API
curl https://kongchang.com/api/v1/knowledge/claims/1000505MCP
get_claim(id=1000505)