We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.
liquid/d1.1Runs on your balance at the prices above. New accounts start with free usage.
Settings
LiquidAI d1.1 is a structured decision model from Liquid AI, positioned as the successor to d1. It sits in the decisions category rather than the general chat category, which tells you what it is for: instead of writing prose, it takes a described situation and returns a choice. Liquid AI serves it as a System One endpoint, meaning the model is built around fast, direct decision output rather than long-form generation.
The model accepts text and images as input, so the state you send can include visual information alongside written description. Its output modality is decisions: you send a state together with typed questions, and it returns a choice. The context window is 65,536 tokens, and the maximum output is 58,982 tokens. That combination of a moderate input budget and a large output budget fits workloads where the incoming state is compact but the set of candidate decisions or the reasoning attached to them can be long.
Practical use cases follow from that shape. Routing and triage are natural fits, where an incoming request or ticket is classified into one of several queues. Selection tasks work the same way: given a state and a set of typed options, the model picks one. Because images are accepted as input, decisions that depend on visual context, such as choosing a category for a submitted photo or judging a document scan, are within scope. Anywhere you would otherwise prompt a chat model to answer with a single label or a structured choice, d1.1 targets that job directly.
You call it through api14's OpenAI-compatible API using the model id "liquid/d1.1". Because the interface follows the OpenAI schema, existing clients and SDKs work without changes; you point them at api14 and set the model field to that id. The model does not list any supported request parameters beyond the standard request shape, so keep your payload to the model id, your messages, and the inputs themselves rather than relying on sampling or formatting controls.
If your task is open-ended writing, summarisation, or multi-turn conversation, a general text generation model is a better fit, since d1.1 returns decisions rather than prose. It is also worth checking that your state and question set fit within the 65,536 token context. For structured choice problems, though, the decision output and image input make it a direct match.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.post("/decisions", { body: {
"model": "liquid/d1.1",
"state": "Customer: I was charged twice for my order and want my money back.",
"questions": {
"team": {
"type": "choice",
"instructions": "Which team should handle this message?",
"criteria": {
"billing": "Payments, refunds, invoices",
"support": "Product questions and bugs",
"sales": "Pricing and new purchases"
}
},
"urgent": {
"type": "noul",
"instructions": "Does this need a reply today?",
"criteria": {
"true": "Money, outage or legal risk",
"false": "Can wait"
}
}
}
} });
console.log(result.answers);
This community has no topics yet. Open the first one about LiquidAI: d1.1.