System one decisionPartner Model

SubQ Snap 1.1

SubQ Snap 1.1 is a multimodal decision model for fast routing, ranking, and classification, delivering structured answers from up to 384K tokens of context.⁠

Talk to an engineer

Model details

Developed by
Subquadratic
Use case
system one decision
Input modalities
TextImage
Output modalities
Structured Decision
License
Subquadratic - Proprietary

SubQ Snap 1.1 is a multimodal decision model built for fast, structured judgments in enterprise agent workflows, including routing, guardrails, ranking, and triage. It accepts text, structured data, and up to four images within a 384K-token context window, returning yes/no probabilities, categorical choices, or numerical scores rather than open-ended text.

SubQ documentation

Input

1curl --request POST \
2  --url https://api.subq.ai/v1/systemone \
3  --header "Authorization: Bearer $SUBQ_API_KEY" \
4  --header 'Content-Type: application/json' \
5  --data @- <<'EOF'
6{
7  "model": "subq-snap-1.1",
8  "state": "Patient with asthma reports a sore throat, fever since yesterday, and shortness of breath walking to the kitchen this morning.",
9  "questions": {
10    "escalate": {
11      "type": "noul",
12      "instructions": "Should a nurse see this chat before the patient books anything?",
13      "criteria": {
14        "true": "Breathing trouble, chest pain, or a flare of a known condition",
15        "false": "Stable symptoms the patient can book for themselves"
16      }
17    },
18    "queue": {
19      "type": "choice",
20      "instructions": "Which team should own this chat?",
21      "criteria": {
22        "nurse_line": "Needs a nurse to assess symptoms today",
23        "primary_care": "Book with the patient's own doctor",
24        "pharmacy": "Refill or medication question only",
25        "front_desk": null
26      }
27    },
28    "acuity": {
29      "type": "score",
30      "instructions": "How soon should someone respond?",
31      "criteria": [
32        "Within a week",
33        "Within a day",
34        "Within an hour"
35      ]
36    }
37  }
38}

JSON output

1{
2    "model": "subq-snap-1.1",
3    "answers": {
4        "escalate": {
5            "type": "noul",
6            "noul": 0.92
7        },
8        "queue": {
9            "type": "choice",
10            "choice": "nurse_line",
11            "probabilities": {
12                "nurse_line": 1,
13                "primary_care": 0,
14                "pharmacy": 0,
15                "front_desk": 0
16            },
17            "confidence": 1
18        },
19        "acuity": {
20            "type": "score",
21            "score": 1.86,
22            "legend": {
23                "0": "Within a week",
24                "1": "Within a day",
25                "2": "Within an hour"
26            },
27            "probabilities": {
28                "0": 0,
29                "1": 0.13,
30                "2": 0.87
31            },
32            "confidence": 0.79
33        }
34    },
35    "usage": {
36        "input_tokens": 477,
37        "output_tokens": 0
38    }
39}