兔子API 兔子API
中

provider

JevAI structured decision models

Direct answerJEV turns business facts into choice, score, and noul signals. On this gateway, use the /v1/systemone endpoint for the Community integration.

Updated · Reviewed

Official positioning

JevAI's Decisions API is built for executable business judgments. Put order, ticket, alert, or review facts in state, then declare questions and criteria in questions. Responses preserve upstream native JSON. Community answers are under data.answers in a code/message/data envelope; AIHubMix and OpenRouter answers are under top-level answers.

Three question types

choice uses a criteria object whose keys are candidate identifiers; the answer includes a choice, probabilities, and confidence. score uses an ordered low-to-high criteria array for risk or priority levels. noul returns a 0–1 probability that a statement is true, useful for escalation or blocking signals. A request can combine these types to classify a ticket and select an action in one call.

Calling through this gateway

The JEV Community channel in the marketplace uses this site's Base URL, site API key, and POST /v1/systemone. The gateway maps the public jev-1.13 name to a Community-compatible Jev identifier and preserves the native response. Configure a JevAI Agent Key only as an upstream channel credential; never use it as a site user token.

{"model":"jev-1.13","state":{"ticket":"Payment API intermittently times out"},"questions":{"urgent":{"type":"noul","instructions":"Should this be escalated immediately?"}}}

For AIHubMix, select the OpenAI channel with https://aihubmix.com as the base URL. For an upstream new-api site, select NewAPI with its site root URL. Both forward to /v1/systemone. Select JevAI for Community (/api/v1/decisions) or OpenRouter for OpenRouter (/api/alpha/decisions). Configure the model endpoint as systemone; the marketplace displays and calls /v1/systemone without inferring Chat Completions from the channel type. Each channel uses its own upstream key.

Boundaries and safety

The current gateway adapter is synchronous and non-streaming, rejects stream, and limits the body to 32 KiB. Probabilities and confidence are decision signals, not factual guarantees; keep human review, authorization checks, and fallbacks in production. Do not place keys, passwords, or unnecessary personal data in state.

Reviewed provider catalog

Latest models by capability

These are provider-catalog models, not a promise of availability on this site. Confirm callable IDs, endpoints, and pricing in the model marketplace.

Text, reasoning, and code

  • jev-1.13 Stable

    A JEV model for structured business decisions with choice, score, noul, and multi-question synchronous judgments.

    Origin: JevAI Community Release date not stated in the cited official source
    Official model catalog ↗ Model details →
Check live availability in the model marketplace

Use cases

  • Ticket triage and routing
  • Risk assessment and human escalation

API protocols

  • /api/v1/decisions
  • /v1/systemone

FAQ

How is JEV different from a chat model?

JEV receives business state and declared questions, then returns structured decision answers. It is designed for judgments and triage rather than long-form conversation.

When should I use choice, score, or noul?

Use choice for one option from named candidates, score for ordered levels, and noul for the probability of a yes judgment. One request may contain all three.

Which path should I call on this gateway?

JEV integrations use /v1/systemone with native JEV JSON. A gateway API key and a JevAI Agent Key are separate credential systems.

Does the gateway stream JEV responses?

The current adapter is synchronous and non-streaming. It rejects stream and caps the request body at 32 KiB.

  • Jev-1.13Call /v1/systemone from the marketplace with state and questions, then read native answers or Community data.answers for triage, routing, and risk decisions.
  • API endpoint and capability matrixConfirm the exact model and enabled endpoint in the marketplace first, then select a protocol here. A registered route only means the gateway recognizes it, not that every model or channel supports it.
  • AI API quickstart: from API key to first requestAn AI API call needs a matching base URL, API key, model ID, and endpoint; start with minimal curl, then add error handling, cost controls, and key security.
  • Common AI API error codes and troubleshootingHandle AI API failures by parsing the HTTP status, provider code, error type, message, retry hints, and request ID together. Fix request, authentication, billing, and hard-quota causes first; retry only transient throttling, network failures, and selected 5xx errors with strict bounds.

Official sources

  1. JevAI API docs Official
  2. AIHubMix JEV docs Official
  3. JevAI Agent Keys Official