Decide first.
Generate second.
Jev is a decision model. It never writes text. You send a situation and typed questions, and it returns one calibrated probability per question in about 100 ms, for $0.042 per million tokens. Put it in front of anything expensive.
Which tier should answer this prompt? The classifier behind a smarter auto.
Free starter credits cover thousands of decisions. Or copy the code below and run it with your own key.
Answers arrive here as probabilities, one card per question, in about 100 ms. No text to parse, nothing invented outside the options you gave.
This is the slot spec 15 left open in `auto`: fast enough to run before first token.
What to do with the numbers
- difficulty ≤ 1 → companyfabric/cheap; ≥ 3 → companyfabric/best; else auto.
- needs_code_model > 0.7 → companyfabric/code.
- Classify once per conversation and cache; re-ask only when the prompt shape changes.
This exact request, in code
curl https://api.companyfabric.com/v1/decisions \
-H "Authorization: Bearer $COMPANYFABRIC_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "typesafe/jev-latest",
"state": "User: Summarise this changelog in three bullets for the release notes.\n\n## 2.14.0\n- Added retry with jitter to the payments client\n- Fixed a race in session refresh\n- Dropped Node 18",
"questions": {
"difficulty": {
"type": "score",
"instructions": "How much reasoning does a correct answer need?",
"criteria": [
"Lookup or reformatting",
"Some judgement",
"Multi-step reasoning",
"Expert analysis or long-horizon planning"
]
},
"tier": {
"type": "choice",
"instructions": "Which model tier should answer?",
"criteria": {
"cheap": "Small fast model",
"balanced": "Mid-tier model",
"frontier": "Best available model"
}
},
"needs_code_model": {
"type": "noul",
"instructions": "Is this primarily a coding task?"
}
}
}'Needs a CompanyFabric key, or add your own TypeSafe key in the BYOK vault and pay them directly at 0% fee. Cost comes back in usage.cost_usd.
One call.
Many answers.
Same key and same balance as the chat models. BYOK with your own TypeSafe key routes at 0% fee. The welcome credit covers thousands of decisions.
Send a state
A ticket, a brief, a transcript, a JSON record. Text or JSON, up to about 32k tokens. It is sent once, whatever you ask about it.
Ask typed questions
noul is a yes/no probability. choice picks one label from criteria you describe. score places on a rubric you define, from zero. Ask several in one call.
Threshold the numbers
Answers are probabilities and distributions, never prose. Act above a threshold, hand to a human below it, and treat a noul near 0.5 as unsure.
Six patterns, ready to run.
Each one is a complete request you can run above, copy as cURL, Node or Python, or download as JSON. Edit the questions to fit your data; the shape stays the same.
Put a decision
in front of the spend.
POST /v1/decisions from your code, or let your agent call the decide tool on our MCP server before every expensive step.