About
Inception Mercury Decide evaluates text, JSON, and images for classification, routing, scoring, and text extraction. It returns typed decisions and probabilities through Inception using named Noul, Choice, Score, and Extract questions. Choice accepts 1–255 options, Score accepts 1–10 levels, and Extract accepts up to 26 fields. Images use embedded data URLs in state or instructions. Requests are synchronous; streaming and tool calling are unavailable.
1. Calling the API
Setup your API Key
To begin using Infron, create an account and generate your API Key.
export INFRON_KEY="YOUR_API_KEY"
Submit a request
Send JSON to POST https://llm.onerouter.pro/v1/decisions using your Infron API key. This is a runnable request example; route availability depends on your deployment.
curl --request POST "https://llm.onerouter.pro/v1/decisions" \
--header "Authorization: Bearer $INFRON_KEY" \
--header "Content-Type: application/json" \
--data '{
"model": "inception/mercury-decide",
"state": "A customer reports being charged twice and asks for a refund today.",
"questions": {
"billing": {
"type": "noul",
"instructions": "Is this request about billing?"
},
"team": {
"type": "choice",
"instructions": "Which team should handle it?",
"criteria": {
"billing": "Payments and refunds",
"technical": "Software support"
}
},
"urgency": {
"type": "score",
"instructions": "How urgent is this request?",
"criteria": [
"Can wait",
"Handle today",
"Immediate emergency"
]
}
},
"provider": {
"only": [
"inception"
],
"allow_fallbacks": false
},
"usage": {
"include": true
}
}'
Response
The following single-answer response illustrates the schema. Its request ID, probabilities, and usage are placeholders, not a captured call.
{
"id": "request_id_example",
"request_id": "request_id_example",
"model": "inception/mercury-decide",
"answers": {
"billing": {
"type": "noul",
"noul": 0.97
}
},
"usage": {
"input_tokens": 100,
"output_tokens": 1
}
}
Routing and usage
provider and usage.include are Infron fields. They are not forwarded to the upstream provider. Set provider.only to ["inception"] and provider.service_tier to "standard" to select this route. usage.include=true includes cost information. Charges use the actual upstream usage.input_tokens; output tokens are free and their actual count is preserved.
Errors
Invalid parameter types or configured limits are rejected before the upstream request. Provider authentication, account, context, and media errors propagate through the existing gateway error handling. The service does not silently truncate the configured question or image arrays. Streaming and asynchronous task polling are unavailable.
2. Schema
Input
| Parameter | Type | Required | Default | Range | Description |
|---|---|---|---|---|---|
model | string | Yes | -- | inception/mercury-decide | Public model ID; upstream selector is mercury-decide. |
state | string / object / array | Yes | -- | See provider contract | Shared content for the questions. |
questions | object | Yes | -- | At least one entry | Named questions; names are retained in the response. |
questions.<id>.type | string | Yes | -- | noul, choice, score, extract | Question type. |
questions.<id>.instructions | string / object / array / null | Conditional | -- | See Question types | What to evaluate; no default is injected. |
questions.<id>.criteria | object / array | For choice and score | -- | See Question types | Options, an ordered rubric, or Extract field definitions (1–26 fields). |
questions.<id>.source | string | Extract, when state is an object | -- | Key of a string field in state | The state field to extract from. |
questions.<id>.threshold | number | No | 0 | 0–1 | Extract: drop spans below this confidence. |
questions.<id>.snap_to_words | boolean | No | true | true, false | Extract: widen spans to whole words. |
questions.<id>.debug | boolean | No | false | true, false | Extract: include extra debugging detail from Inception. |
provider | object | No | -- | -- | Platform routing options. |
usage.include | boolean | No | false | true, false | Include platform cost details. |
Question types
Noul returns a yes probability. Choice accepts 1–255 options; Score accepts 1–10 ordered levels, indexed from zero. Choice requires an option map; Score requires an array. Instructions and criteria describe the task according to the upstream contract; Noul requires a meaningful question through instructions or criteria. Extract questions accept up to 26 named fields; provide instructions, criteria, or both. Extract answers contain spans and not_stated. When state is an object, set source to the key of the string field to extract from. Requests that contain Extract questions cannot include images and are limited to 32,768 tokens. The official model table does not publish a context limit.
Media inputs
Embed PNG, JPEG, WebP, or GIF data-URL image parts in state or questions.<id>.instructions, not criteria. Up to 16 images of at most 10 MiB each are accepted, subject to the 20 MiB whole-request limit. Remote URLs are unsupported. Images are not accepted in requests that contain Extract questions.
Output
| Parameter | Type | Range | Description |
|---|---|---|---|
id, request_id | string | -- | Platform request ID. |
model | string | -- | Public model ID. |
provider, service_tier | string | -- | Selected route. |
answers | object | Request question IDs | Typed upstream answers. |
answers.<id>.type | string | noul, choice, score, extract | Answer type. |
answers.<id>.noul | number | 0–1 | Yes probability. |
answers.<id>.choice | string | Choice keys | Selected option. |
answers.<id>.score | number | 0 to levels minus 1 | Probability-weighted level. |
answers.<id>.confidence | number | 0–1 | Choice/Score confidence. |
answers.<id>.probabilities | object | Values 0–1 | Probability distribution. |
answers.<id>.legend | object | -- | Score level descriptions. |
answers.<id>.spans | object | -- | Extract: per field, the matched spans as text, start, end, and confidence. |
answers.<id>.not_stated | object | Values 0–1 | Extract: per field, the probability that the value is not stated. |
usage.input_tokens, usage.output_tokens | integer | Non-negative | Actual upstream usage. |
cost | number | USD | Charge details when requested. |
cost_details | object | -- | Billing breakdown when requested. |
3. Detailed Pricing
| Model | Billing unit | Input price | Output price |
|---|---|---|---|
| Inception Mercury Decide | Per 1 million tokens | $0.02 (list $0.04) | Free |
Inception currently offers Mercury Decide at 50% off its $0.04 list price, with no published end date; Infron passes the $0.02 price through and follows Inception if the promotion ends. Cached input is free. Charges use the actual upstream usage.input_tokens; output tokens are free and their actual count is preserved. Official pricing.