
Jev AI
A decision model and API for software teams: submit a state and typed questions, and get typed decisions with probability that your code can use directly.
About Jev AI
Jev AI's core use case is decision-making: developers submit a state and a set of typed questions, and the request returns typed results with probability and confidence that can be written directly into application logic. Each question asks exactly one clearly scoped thing, and multiple questions are evaluated in parallel against the same state, so there is no need to chain multiple requests just to split up decisions. Inputs currently support text, JSON objects, and text arrays; images, audio, and video are not yet supported, and accuracy for non-English input must be verified separately. Pricing is available monthly, yearly, or as a one-time purchase; one-time plans do not auto-renew, and the credits they include remain available long-term after purchase.
Best for
Best use cases
When your app needs fast text classification or routing, get results with probability in a single request
When you need to score a state and drive business logic based on the score
Insert content safety checks into a workflow and use confidence to decide whether human review is needed
Validate real workflows in the playground first, then integrate into production with an API key
Features
Typed decision output
Send one state and a set of typed questions, and get decision results your code can use directly, including probability and confidence.
Parallel multi-question evaluation
Multiple questions are evaluated in parallel against the same state, with no need to chain calls to split up decisions.
Choice and Score question types
The Choice type is for classifying or routing among options and returns the option, probability, and confidence; the Score type is for scoring.
REST API and SDK
Verify your workflow in the console, then create an API key; production requests call the API with three top-level fields (state, model, questions).
Online playground
Validate real scenarios first in the playground on the website, then integrate via the API.
Pros
- Output is a typed decision with probability and confidence that can be used directly in application logic
- Multiple questions are evaluated in parallel against the same state, with no need for chained calls
- Offers an online playground and official documentation, so you can verify before integrating
- Credits in one-time plans remain available long-term after purchase
Cons
- Currently only supports text, JSON objects, and text arrays; image, audio, and video inputs are not yet supported
- Accuracy for non-English input must be verified separately on your own
- Official notice states that, due to limited resources, plans may be removed at any time to ensure service quality and speed for paying users
- Probability and confidence are signals for automation and do not guarantee business-level accuracy; high-risk operations require higher thresholds or human review
Pricing Plans
Starter
- 100,000 credits, never expire
- 1 workspace
- Choice, Score, and noul questions
- Typed decision output
- Probability and confidence results
- Online playground
- API key management
- Email support
Pro
- 1,000,000 credits, never expire
- Unlimited workspaces
- Choice, Score, and noul questions
- Typed decision output
- Multiple questions per request in parallel
- API access
- Usage and request history
- Priority support
Enterprise
* Pricing information is for reference only. Please refer to the latest announcement on Jev AI's official website.