elephant-1
elephant-1A 9B decision model that turns structured context into calibrated choice, yes/no, and scoring probabilities in a single forward pass.
Model details
View repositoryelephant-1 is a 9B decision model that turns 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.
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), elephant-1 is the top 9B model on the Jev Decision Index 0.4 and Artificial Analysis Decision Index.
elephant-1 is the leading 9B model on the Jev Decision Index 0.4 with a score of 62.6. On the Artificial Analysis Decision Index, elephant-1 score 65.2.1curl --fail-with-body \
2 -X POST "https://inference.baseten.co/v1/systemone" \
3 -H "Authorization: Api-Key $BASETEN_API_KEY" \
4 -H "Content-Type: application/json" \
5 -d '{
6 "model": "decision-models/baseten-elephant-9b-latest",
7 "state": {
8 "message": "I was charged twice. Please refund the duplicate charge."
9 },
10 "questions": {
11 "route": {
12 "type": "choice",
13 "instructions": "Which team should handle this request?",
14 "criteria": {
15 "billing": "Payments, charges, and refunds",
16 "support": "Technical troubleshooting"
17 }
18 }
19 }
20 }'
