Give it context
Send the document, request, or state your software is working with.
Intelligence, in the loop
Decision models that turn context into choices, scores, and probabilities. Built for the moments when your code needs to make a judgment.
Three model sizes. One simple API.
“I upgraded yesterday, but my account still shows the free plan. My card was charged. Can you help?”
Which team should handle this?
ticket.assign(answer.choice)YOUR CODEA small building block
Route a request, check a condition, or score a result. Compose decisions into workflows you control.
Send the document, request, or state your software is working with.
Ask one or more typed questions. You choose the options and the rubric.
Read the probabilities. Set thresholds, take action, or send uncertain cases for review.
Right-sized intelligence
Per million input tokens.
No output charges.
Start small. Evaluate focused, everyday decisions.
$0.021/ 1M input tokens
decision-model-tinyA starting point for a wide range of workflows.
$0.042/ 1M input tokens
decision-model-smallOur largest model. Compare it on your harder decisions.
$0.084/ 1M input tokens
decision-model-largeAccess is by approval. Start with a $50 organization allowance; usage pauses when it is exhausted. No automatic charges.
From experiment to application
Explore questions in the playground, compare models on your examples, then use the same request in your application. Manage your team, API keys, and usage in one console.
Make your first request{
"model": "decision-model-small",
"state": "A customer was charged twice.",
"questions": {
"team": {
"type": "choice",
"instructions": "Which team should help?",
"criteria": {
"billing": "Payments and invoices",
"technical": "Product issues"
}
}
}
}Build with Spinda