d1.2 overview
Decision example
null
| Question | Result |
|---|---|
| Damaged item | P(yes): 0.5454 |
| Team | returns (confidence 0.1194) |
| Urgency | 0.6565 / 2 (confidence 0.0153) |
Actual model response from Nov 17, 2026.
View request and response
Request
{
"model": "liquid/d1.2",
"questions": {
"damaged_item": {
"type": "noul",
"instructions": "Does the customer report an item that arrived damaged?"
},
"team": {
"type": "choice",
"instructions": "Which team should handle this request?",
"criteria": {
"returns": "Damaged products, replacements and returns",
"billing": "Payments, invoices and charges",
"technical": "App errors and integrations"
}
},
"urgency": {
"type": "score",
"instructions": "Rate how urgently the request needs attention.",
"criteria": [
"Routine follow-up",
"Customer is blocked",
"Immediate safety risk"
]
}
}
}Response
{
"model": "d1.2",
"answers": {
"damaged_item": {
"type": "noul",
"noul": 0.5453985179817601
},
"team": {
"type": "choice",
"choice": "returns",
"probabilities": {
"returns": 0.4129030975264723,
"billing": 0.286867870451208,
"technical": 0.30022903202231976
},
"confidence": 0.11935464628970849
},
"urgency": {
"type": "score",
"score": 0.656483695471622,
"legend": {
"0": "Routine follow-up",
"1": "Customer is blocked",
"2": "Immediate safety risk"
},
"probabilities": {
"0": 0.46121807764781225,
"1": 0.42108014923275344,
"2": 0.11770177311943425
},
"confidence": 0.015274456792567004
}
},
"usage": {
"input_tokens": 128,
"output_tokens": 0
}
}d1.2 by Liquid AI
Liquid AI built d1.2 as its default decision model. It reads text and images and returns typed answers without generating output tokens. Against the earlier decision models, Liquid AI says d1.2 adds selecting multiple applicable options.
d1.2 specs
- Model ID
liquid/d1.2- Developer
- Liquid AI
- Context window
- 65,536 tokens
- Input
- Image, Text
- Output
- Text
- Endpoint
POST https://api.aimlapi.com/v1/decisions
Ask many questions about one input
A request carries a map of questions, from 1 to 128 of them. Every question key comes back in the answers field with a result of the type you asked for.
Each result is a number or a key you can read directly. A yes/no answer is a probability between 0 and 1. A choice answer is an option key with a confidence and a probability for every option.
Pick the question type that matches the decision
Use a yes/no question for gates such as a policy check or an LLM-as-judge verdict. The answer is the probability of yes, so you set the threshold. You can describe what true and false mean for the question in text or as a JSON object.
Use a choice question for routing and labeling. It takes 2 to 255 options, and an option description can be text, a JSON object or null when the key says it all. Turn on the multiple setting when several options can apply at once.
Where d1.2 fits
Liquid AI recommends d1.2 for classification, routing, scoring, extraction and grounding when you prefer a faster hosted decision model. Put it in front of a larger generative model to decide which prompt, queue or reviewer an input goes to. Check the probabilities on your own edge cases before you automate the routing.
Use cases
Route incoming support tickets to the right queue with a choice question.
Gate content with yes/no questions and act on the returned probability.
Rate answers or documents on an ordered scale as an LLM-as-judge check.
Inspect product photos and tag every defect type that applies with the multiple setting.
Capabilities
- Vision
d1.2 API pricing
| Usage | Price | Unit |
|---|---|---|
| Input | $0.02 | 1M tokens |
| Output | $0 | 1M tokens |
Prices from the AI/ML API price list, checked on Nov 17, 2026.
How to use the d1.2 API
View docs (opens in a new tab)Build with your agent
Give your agent the model details, code and documentation.
1Get your API key
Create API keyexport AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"2Make your first request
curl 'https://api.aimlapi.com/v1/decisions' \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "liquid/d1.2",
"questions": {
"damaged_item": {
"type": "noul",
"instructions": "Does the customer report an item that arrived damaged?"
},
"team": {
"type": "choice",
"instructions": "Which team should handle this request?",
"criteria": {
"returns": "Damaged products, replacements and returns",
"billing": "Payments, invoices and charges",
"technical": "App errors and integrations"
}
},
"urgency": {
"type": "score",
"instructions": "Rate how urgently the request needs attention.",
"criteria": [
"Routine follow-up",
"Customer is blocked",
"Immediate safety risk"
]
}
}
}'import os
import requests
headers = {
"Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
"Content-Type": "application/json",
}
payload = {
"model": "liquid/d1.2",
"questions": {
"damaged_item": {
"type": "noul",
"instructions": "Does the customer report an item that arrived damaged?"
},
"team": {
"type": "choice",
"instructions": "Which team should handle this request?",
"criteria": {
"returns": "Damaged products, replacements and returns",
"billing": "Payments, invoices and charges",
"technical": "App errors and integrations"
}
},
"urgency": {
"type": "score",
"instructions": "Rate how urgently the request needs attention.",
"criteria": [
"Routine follow-up",
"Customer is blocked",
"Immediate safety risk"
]
}
}
}
response = requests.post("https://api.aimlapi.com/v1/decisions", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())const response = await fetch("https://api.aimlapi.com/v1/decisions", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "liquid/d1.2",
"questions": {
"damaged_item": {
"type": "noul",
"instructions": "Does the customer report an item that arrived damaged?"
},
"team": {
"type": "choice",
"instructions": "Which team should handle this request?",
"criteria": {
"returns": "Damaged products, replacements and returns",
"billing": "Payments, invoices and charges",
"technical": "App errors and integrations"
}
},
"urgency": {
"type": "score",
"instructions": "Rate how urgently the request needs attention.",
"criteria": [
"Routine follow-up",
"Customer is blocked",
"Immediate safety risk"
]
}
}
}),
});
if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());d1.2 FAQ
How much does d1.2 cost?
d1.2 is billed at $0.02 / 1M tokens for input. There are no output-token charges.
How many questions can one request hold?
A request can hold 1 to 128 questions. Each question key returns its own typed result in the answers field.
How do I see how sure the model is?
A yes/no answer is the probability of yes. Choice and score answers add a calibrated confidence and a probability for each option or level.
What does d1.2 add over the previous decision models?
Liquid AI lists selecting multiple applicable options among the additions. In a choice question you enable it with the multiple setting.
What is the d1.2 model ID?
Use liquid/d1.2 as the model parameter. Requests go to https://api.aimlapi.com/v1/decisions with your AI/ML API key.