Learn the Jev API. Put agent decisions to the test.

Follow practical guides for model routing, tool-call review and evidence checks. Then test a native Jev request in our independent playground with your own OmniaKey account.

Open live playground Start with the API guide

Connect Jev AI

api.omniakey.com

Call the OmniaKey Jev API

/v1/alpha/search

curl -X POST "https://api.omniakey.com/v1/alpha/search" \
  -H "Authorization: Bearer ${OMNIAKEY_API_KEY}" \
  -H "Content-Type: application/json" \
  --data '{
  "model": "jev-latest",
  "state": {
    "task": "Review a complex customer dispute with 100k context and tool use",
    "candidates": [
      {
        "id": "fast-model",
        "description": "Fast general model with 32k context",
        "cost": "low",
        "latency": "low"
      },
      {
        "id": "reasoning-model",
        "description": "Strong reasoning, 200k context, and tool use",
        "cost": "high",
        "latency": "medium"
      }
    ],
    "priorities": [
      "quality",
      "context",
      "tool_use",
      "cost"
    ],
    "constraints": [
      "Customer data must remain within approved tools"
    ],
    "stakes": "high"
  },
  "questions": {
    "model": {
      "type": "choice",
      "instructions": "Select the best approved candidate using the stated quality, context, tools, cost, latency, and stakes. Do not assume undocumented capabilities. Choose escalate when no candidate is adequate or evidence is insufficient.",
      "criteria": {
        "fast-model": "Choose this approved model if its documented capabilities satisfy the task, priorities, constraints, and stakes: {\"id\":\"fast-model\",\"description\":\"Fast general model with 32k context\",\"cost\":\"low\",\"latency\":\"low\"}",
        "reasoning-model": "Choose this approved model if its documented capabilities satisfy the task, priorities, constraints, and stakes: {\"id\":\"reasoning-model\",\"description\":\"Strong reasoning, 200k context, and tool use\",\"cost\":\"high\",\"latency\":\"medium\"}",
        "escalate": "No approved candidate can safely satisfy the task, or the provided evidence is insufficient to select one."
      }
    }
  }
}'

Agent setup

Use your OmniaKey API key to send a native Jev request and receive model answers.

Create an access key

Jev model guide

Explore a decision, then inspect the evidence.

Jev is a decision model developed by TypeSafe AI. This independent playground helps you define the state and questions, inspect the exact request, and review the answer returned through your OmniaKey account.

Try a native Jev request

Model basics Jev AI for agents Jev model guide Jev API guide Playground features Workflow presets

DimensionJevGenerative LLM

OutputTyped choices, probabilities, and scoresGenerated text, code, and tool arguments

Best useRouting, classification, scoring, and gatesPlanning, explanation, coding, and creation

UncertaintyCalibrated probabilities are part of the responseConfidence handling depends on the workflow

Structured requests

Test the input and inspect the answer before adding a workflow step.

Jev AI pairs Jev with validated inputs, editable workflow presets and structured answers through OmniaKey. The agent keeps its existing model, tools, permissions, and approval rules.

Explore API setup

Practical Jev API guides

Understand an example. Then try your own.

Explore all 9 guides →

Start here

Jev API quickstart with OmniaKey

Make your first Jev API request with an OmniaKey key. Follow curl and Node.js examples, inspect the response, and troubleshoot authentication errors.

Read guide Routing

LLM model routing with Jev: a practical example

Build a bounded LLM router with Jev. Describe approved models, compare task constraints, handle escalation and evaluate cost against a fixed-model baseline.

Read guide Tool review

AI agent tool-call guardrails with Jev

Review proposed agent tool calls with Jev using action choices and a risk scale. Keep authorization, approval thresholds and execution checks in your code.

Read guide

Decisions your agent should not make from instinct alone.

Jev AI makes Jev decisions available through editable workflow presets and a native API playground.

model-route

Smart model routing

Choose from the models you allow using task stakes, quality, cost, latency, context, and tool-use needs. Escalate when no candidate is safe enough.

Input: task + candidates
Signals: quality / cost / latency
Output: model + escalate

tool-guard

Tool-call protection

Inspect the proposed action, side effects, safeguards, reversibility, and your policy before an agent changes external state.

allow / confirm / review / deny

Research Guard

Judge whether the available evidence is sufficient, contradictory, stale, or ready to cite.

Completion Reviewer

Decide whether a task is truly done, needs one more check, or should return to implementation.

Custom decision contracts

Define your own choice, yes-or-unknown, and scored questions through the generic Decisions API.

One Jev decision layer, four ways to use it.

Jev AI brings together a live decision playground, native request examples and integration guides for agent workflows.

Native request examples

https://api.omniakey.com/v1/alpha/search

Copy a REST or server-side JavaScript example, set your OmniaKey key, and send the same request from your own application.

View examples

OmniaKey Jev API

https://api.omniakey.com/v1/alpha/search

Send native Jev state and questions through OmniaKey for model routing, tool protection, research, task routing, completion, or custom decisions.

View API docs

Editable workflow presets

/playground?preset=model-route

Start from model routing, tool review, task routing, research or completion checks. Inspect the input and the resulting native request before you run it.

Explore the workflow

Online Jev playground

/playground

Edit preset inputs, call Jev through OmniaKey, and inspect the model’s actual choices, probabilities, scores and token usage.

Start debugging

Live Jev API debugger

Inspect the decision before your agent acts.

Run Jev through OmniaKey with a workflow preset or your own native decision contract. Edit the JSON input and inspect the model’s actual answers, probabilities, and token usage. Your OmniaKey API key stays in memory and is never saved.

Read the guide for this preset →

Choose a workflow preset and provide the task context. The preset defines bounded Jev questions, evaluates your input through OmniaKey, and returns the model’s original answers. Open the native request preview to inspect what will be sent.

OmniaKey API key

Used only for this request. The page does not persist the key.

Request endpoint

https://api.omniakey.com/v1/alpha/search

All presets use OmniaKey’s native Jev endpoint with the jev-latest model.

Workflow preset input

{
  "tool": "issue_customer_refund",
  "action": "Refund USD 680 after a disputed duplicate charge",
  "arguments_summary": [
    "order_id=ord_7429",
    "amount_usd=680"
  ],
  "side_effects": [
    "Moves funds",
    "Changes the order payment state"
  ],
  "safeguards": [
    "Customer identity and duplicate charge verified"
  ],
  "policy": [
    "Refunds above USD 500 require human approval"
  ],
  "reversibility": "partially_reversible"
}

View the native request sent to OmniaKey

{
  "model": "jev-latest",
  "state": {
    "tool": "issue_customer_refund",
    "action": "Refund USD 680 after a disputed duplicate charge",
    "arguments_summary": [
      "order_id=ord_7429",
      "amount_usd=680"
    ],
    "side_effects": [
      "Moves funds",
      "Changes the order payment state"
    ],
    "safeguards": [
      "Customer identity and duplicate charge verified"
    ],
    "policy": [
      "Refunds above USD 500 require human approval"
    ],
    "reversibility": "partially_reversible"
  },
  "questions": {
    "action": {
      "type": "choice",
      "instructions": "Evaluate the proposed tool call using policy, side effects, safeguards, authorization, and reversibility. Treat policy thresholds and missing approvals as constraints. This evaluation does not grant permission to execute anything.",
      "criteria": {
        "allow": "The action is authorized, within policy, and supported by adequate safeguards.",
        "confirm": "Ask the user to confirm intent or missing details before proceeding.",
        "review": "A policy approval or human review is required before this action can proceed.",
        "deny": "The action violates policy, lacks required authority, or must not proceed."
      }
    },
    "risk": {
      "type": "score",
      "instructions": "Rate the financial, operational, and policy risk of the proposed action.",
      "criteria": [
        "Low: routine and within policy.",
        "Moderate: limited, reversible effects.",
        "High: material effects or approval requirements.",
        "Critical: unsafe, unauthorized, or irreversible effects."
      ]
    }
  }
}

Create an access key

Structured response

answers + probabilities + usage

The typed Jev response will appear here.

Run the request to inspect the original Jev answers, confidence, probability distributions, resolved model version, and token usage returned through OmniaKey.

Decision boundaries

Use Jev where the workflow needs an answer, not another paragraph.

Each Jev AI preset turns compact workflow state into a bounded decision your agent can inspect and apply within its existing permissions.

Explore all use cases

Model routing

Choose the model that fits the task

Compare only the models you allow against quality, context, tool use, cost, latency and task stakes.

Workflow question: Which approved model can handle 100k context and tool use without wasting cost?
Example decision: reasoning-model

Read the use case Try this preset

Tool-call guardrails

Review an action before it changes external state

Evaluate side effects, safeguards, policy thresholds and reversibility before an agent calls a consequential tool.

Workflow question: Can the agent issue this high-value refund without human confirmation?
Example decision: review · 0.91

Read the use case Try this preset

Task routing

Select the next workflow path

Route ambiguous work to quick execution, deeper review, decomposition or a blocked state using explicit evidence.

Workflow question: Should the agent proceed, review deeply, split the task or stop?
Example decision: review_deep

Read the use case Try this preset

Research verification

Check whether the evidence supports the claim

Judge evidence coverage, contradictions, source quality and stakes before an agent cites or acts on research.

Workflow question: Is the current evidence sufficient to support this policy claim?
Example decision: insufficient

Read the use case Try this preset

Completion review

Keep unfinished work from being reported as done

Compare the objective, completed work, verification and known gaps before closing the task.

Workflow question: Is the refund task complete if the charge is identified but no refund was issued?
Example decision: verify_more

Read the use case Try this preset

The value is not another model endpoint.

Jev supplies structured judgment. Jev AI lets you test those decisions and inspect the original model answers before using them in your agent.

Decision engine

Jev evaluates

A System One model selects among explicit options and returns probabilities and scores.

Jev AI

Operational decision layer

Contract

Typed inputs and bounded outputs for each decision.

Connect

Editable inputs, native request previews, and REST examples.

Trigger

Teach agents which boundary needs a second judgment.

Action policy

Map an answer to proceed, confirm, review, or stop.

Execution layer

Your agent acts

The agent keeps its model, tools, permissions, and human approval boundaries.

Agent integration

Connect once. Add judgment at every decision boundary.

Pick a preset, inspect the native request, then run it with your own OmniaKey key. Decide how your application should use the result.

Prepare your OmniaKey key

Create a key in your OmniaKey account. For your own server integration, keep the key in an environment variable.

https://omniakey.com/dashboard/tokens

Choose a workflow preset

Edit the task input for routing, tool review, research or completion checks, then inspect the generated questions.

/playground?preset=model-route

Evaluate with Jev

Send compact task state and bounded questions, then receive choices, probabilities, and scores.

choice + noul + score

Apply your policy

Let the workflow proceed, request confirmation, route for review, or stop without expanding authority.

allow / confirm / review / deny

Make the first guarded decision.

Start with a small test request. Inspect the returned answer in the playground, then copy the REST example into your own workflow.

Create an OmniaKey API key from your account settings.
Open the playground and choose a preset, or send the REST request shown here directly to OmniaKey.
Review the answer, model version and token usage before connecting your own automation.

Create an access key Read OmniaKey docs

OmniaKey REST request

jev-latest · /v1/alpha/search

curl -X POST "https://api.omniakey.com/v1/alpha/search" \
  -H "Authorization: Bearer ${OMNIAKEY_API_KEY}" \
  -H "Content-Type: application/json" \
  --data '{
  "model": "jev-latest",
  "state": "This is a synthetic integration test. No external actions are requested.",
  "questions": {
    "is_test": {
      "type": "noul",
      "instructions": "Is the input an integration test?"
    }
  }
}'

Set OMNIAKEY_API_KEY in your terminal environment before running. OmniaKey manages your key and billing; this site does not sell API credits.

Jev AI questions, answered.

Learn what this playground sends, how to use the results, and where your account is managed.

Jev is a decision model developed by TypeSafe AI. It returns structured choices, probabilities and scores. Jev AI is an independent playground for trying this model through OmniaKey; this website is not the model developer or an official OmniaKey website.

No. The Jev model is designed for bounded decisions inside software: classify, route, score, verify or choose among options defined by the caller. Use a generative model for writing and planning, then use Jev at decision boundaries.

Start with model routing, tool-call review, task routing, research checks or completion review. Each preset turns a small JSON input into a native Jev request. The custom preset lets you define your own state and questions.

You can call OmniaKey directly. Jev AI adds editable workflow presets, input validation and a native request preview so you can inspect the model’s structured answers before integrating them into your workflow.

Accounts, API keys, billing and provider support are managed by OmniaKey. Jev AI does not issue keys or sell API credits. Review OmniaKey’s current terms, supported regions and model availability before using your account here.

No. The playground returns structured answers, probabilities and scores. Your agent or workflow decides what happens next, within its existing permissions.

When you run a request, your API key and the JSON you entered pass through the Jev AI server to OmniaKey. The application does not persist or log your key or request body; hosting and provider services may retain operational logs under their own policies. The page keeps the key in memory until it is cleared or the page closes. Do not include secrets or private customer data in the task input.

Yes. Use your OmniaKey API key in the playground, inspect the native request, then review the returned model version, answers and token usage. Calls use your provider account and may incur charges according to OmniaKey’s pricing.

Give your agent a decision boundary.

Use your OmniaKey key to try a small request, review the response, and decide how it fits your workflow.

Create an access key

Learn the Jev API. Put agent decisions to the test.

Explore a decision, then inspect the evidence.

Test the input and inspect the answer before adding a workflow step.

Understand an example. Then try your own.

Jev API quickstart with OmniaKey

LLM model routing with Jev: a practical example

AI agent tool-call guardrails with Jev

Decisions your agent should not make from instinct alone.

Smart model routing

Tool-call protection

Research Guard

Completion Reviewer

Custom decision contracts

One Jev decision layer, four ways to use it.

Native request examples

OmniaKey Jev API

Editable workflow presets

Online Jev playground

Inspect the decision before your agent acts.

Use Jev where the workflow needs an answer, not another paragraph.

Choose the model that fits the task

Review an action before it changes external state

Select the next workflow path

Check whether the evidence supports the claim

Keep unfinished work from being reported as done

The value is not another model endpoint.

Jev evaluates

Operational decision layer

Your agent acts

Connect once. Add judgment at every decision boundary.

Prepare your OmniaKey key

Choose a workflow preset

Evaluate with Jev

Apply your policy

Make the first guarded decision.

Jev AI questions, answered.

What is Jev?

Is the Jev model a chatbot or generative LLM?

What can I test here?

Why not call the Jev endpoint directly?

Where do I sign in and manage billing?

Can Jev AI execute actions on its own?

What context will be sent to Jev?

Is Jev AI available now?

Give your agent a decision boundary.