decision-models/baseten-joey-2b-latest
A 2B decision model that turns structured context into calibrated choice, yes/no, and scoring probabilities in a single forward pass.
joey-1 is a 2B decision model fine-tuned from Qwen3.5-2B to turn structured context into structured answers. Given a state and one or more typed questions, it returns answer probabilities and confidence in a single forward pass without generating free-form text that an application must parse.
| Attribute | Value |
|---|---|
| Parameter Size | 2B |
| Context Length | 16k |
| Input Modalities | Text, Image |
| Output Modalities | Structured Decisions |
Use it for tasks such as routing requests, classifying cases, answering yes/no checks, or scoring against a rubric. It supports choice questions, yes/no questions, and ordered scoring, returning probabilities your application can use to select an answer or set confidence thresholds.
On release (October 2026), joey-1 is the top 2B model on the Jev Decision Index 0.4.
Input
curl --fail-with-body \
-X POST "https://inference.baseten.co/v1/systemone" \
-H "Authorization: Api-Key $BASETEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "decision-models/baseten-joey-2b-latest",
"state": {
"message": "I was charged twice. Please refund the duplicate charge."
},
"questions": {
"route": {
"type": "choice",
"instructions": "Which team should handle this request?",
"criteria": {
"billing": "Payments, charges, and refunds",
"support": "Technical troubleshooting"
}
}
}
}'