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-latestRuns on your balance at the prices above. New accounts start with free usage.
Settings
LiquidAI: d1 Latest is a model from LiquidAI that sits in the decisions category. It is a routing alias rather than a fixed snapshot: the model always redirects to the latest model in the d1 family, so the underlying version can change over time while the identifier stays the same. That makes it a convenient choice when you want to stay on the current d1 release without updating your integration each time a new version ships.
The model accepts text and images as input and produces decisions as output. Its context window is 65,536 tokens, and the maximum output is 58,982 tokens. Because the output modality is decisions rather than free form text, the model is built to return a chosen option or verdict rather than a prose answer. The combination of image input and a 65k context means a single request can carry a substantial amount of textual material alongside visual content before a decision is made.
Practical use cases follow directly from that shape. Classification and routing tasks fit well, such as deciding which category a support ticket belongs to or which team should handle a request. Approval and gating logic is another natural fit, where the model reviews a request and returns a decision. Because images are accepted, decisions that depend on visual input also work, for example reviewing a screenshot or a photo together with accompanying text and returning a verdict. Any workflow where you need a discrete outcome rather than generated prose is a candidate.
You call the model through api14's OpenAI-compatible API using the model id "~liquid/d1-latest". Because the interface follows the OpenAI format, existing OpenAI client libraries and tooling work without changes; you point your client at the api14 endpoint and pass the model id in the request. The model supports no additional request parameters beyond the standard ones, so there are no provider-specific options to configure. If you rely on sampling controls or other tuning knobs, they are not available here, and the request should be kept to the standard fields.
One thing to keep in mind is that the alias tracks the latest d1 model, so behaviour can shift when LiquidAI updates the family, which matters if you need reproducible results across time. Pinning to a specific d1 version would be the alternative in that case. The decisions output modality also means this model is not the right fit when you need generated text, summaries or conversational replies; a text generation model would serve those needs better.
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-latest",
"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 Latest.