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.

New here? See how it works A 1-minute tour of the playground, models and your API key.

d1-ultra vs Jev 1.14: API price, context & features · api14

LiquidAI: d1-ultra vs TypeSafe: Jev 1.14

A side-by-side comparison of LiquidAI: d1-ultra ($0.15 / 1M tokens) and TypeSafe: Jev 1.14 ($0.15 / 1M tokens) using current USD prices and capabilities from the api14 catalog. Both run through the same OpenAI-compatible API.

Specifications

	LiquidAI: d1-ultra	TypeSafe: Jev 1.14
Provider	LiquidAI	TypeSafe
Input	$0.15 / 1M tokens	$0.15 / 1M tokens
Cache read	$0.15 / 1M tokens	-
Context window	32,768	262,144
Max output	29,491	235,929
Input	text, image	text, image
Output	decisions	decisions
Tool calling
Structured output
Reasoning control
Added	Oct 15, 2026	Oct 14, 2026

Key differences

Both have about the same input price.
TypeSafe: Jev 1.14 reads 8.0× more context (262,144 tokens).
A request with 1,000 input and 500 output tokens costs $0.000147 with LiquidAI: d1-ultra and $0.000147 with TypeSafe: Jev 1.14.

In depth

LiquidAI: d1-ultra is a model from LiquidAI, and TypeSafe: Jev 1.14 is a model from TypeSafe. Both sit in the decisions category, meaning their output is not free form prose but a decision, so you should treat them as components that return a choice rather than as chat models that write text back. LiquidAI released d1-ultra on 2026-10-15, one day after TypeSafe released Jev 1.14 on 2026-10-14. If your architecture already assumes a decisions endpoint, either model can slot into that role.

The two are close on input handling and far apart on how much they can read at once. Both accept text and image input, and both emit decisions, so neither has a modality advantage over the other. The real split is context length: d1-ultra supports 32768 tokens, while Jev 1.14 supports 262144 tokens, which makes Jev 1.14 the one with the bigger context by a wide margin. Neither model lists any supported parameters, so you cannot tune sampling behaviour, temperature or similar knobs through the request body. On price, the two are about the same for input, and there is no meaningful difference recorded for output cost, so cost is not the axis that should decide this for you.

Pick based on how much material the decision has to be made over. If the decision depends on a large body of text, a long transcript, a big document set, or many images plus their surrounding context, Jev 1.14 is the safer choice because its context window is much larger and you are less likely to truncate the evidence the decision rests on. If the decision is narrow and the relevant input fits comfortably inside 32768 tokens, d1-ultra is a reasonable pick and there is no cost penalty for choosing it, since input pricing is about the same. In both cases, remember that the model returns a decision, so the surrounding logic, the allowed set of outcomes and the validation of that output stay on your side.

Calling either one through api14 works the same way because the API is OpenAI compatible. You point your existing client at api14, keep your usual request shape, and change only the model id field between "liquid/d1-ultra" and "typesafe/jev-1.14". Everything else in the call, the messages, the text and image parts, and the response handling, stays as it is. That means you can build against one id, then switch the id to compare the other on your own inputs without rewriting integration code. Since neither model exposes supported parameters, do not send tuning fields expecting them to take effect.

There is no case in the facts given where one model is clearly the better pick on a dimension other than context. They tie on input price, they match on input and output modalities, and neither offers tools, structured output or reasoning as a distinguishing feature. So the decision reduces to a single question: does your input fit in 32768 tokens? If yes, either id will do and you can choose on other grounds. If no, use "typesafe/jev-1.14" and keep "liquid/d1-ultra" for the smaller, self contained decisions where its shorter context is not a constraint.

Switch with one string

Both models use the same endpoint and request shape. Change the model value from "liquid/d1-ultra" to "typesafe/jev-1.14" and prompts, streaming and tools keep working.

LiquidAI: d1-ultra details & playground →TypeSafe: Jev 1.14 details & playground →

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": "typesafe/jev-1.14",
  "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);

LiquidAI: d1-ultra vs TypeSafe: Jev 1.14

Specifications

Key differences

In depth

Switch with one string

More comparisons