Jev (TypeSafe)

edenai fast tier Decision model default for Decision model

Typed decisions about text or JSON — yes/no with a probability, a choice among labelled categories with per-label probabilities, or a score on an ordered rubric — in one sub-second pass. Built for routing, moderation, classification, scoring and validation where a calibrated probability matters and prose does not. Text/JSON input only, ~64k-token context.

metered
Published price
jev
species_name
2026-10-10
Verified since
Call it

One request, the price in the response.

Pass species_name to pin this model; omit it and Fire uses the default for the endpoint. Every response carries meta.price.usd and meta.call_id.

Get your API key → Full API reference
curl -X POST https://fire.prosaga.net/v1/decide \
  -H "Authorization: Bearer $FIRE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"species_name":"jev",
      "state":{"message":"My invoice is wrong"},
      "questions":[{"id":"intent","type":"choice","options":["billing","support","sales"]}]}'

Strengths

decisionclassificationscoringstructured-outputlow-latencycalibrated-probabilities

Modality

In: text  ·  Out: text

Lineage

Family typesafe · genus jev · served by edenai

Capabilities

input
["text","json"]
modality
decision
question types
["noul","choice","score"]
max state tokens
32000
max context tokens
64000

Call Jev (TypeSafe) with one key.

Free to sign up. Pay only when you top up.

Get your API key → Back to the catalog