Decart

The Decart community adapter calls Decart image and video models from generateImage() and generateVideo(). Decart maintains it.

1. Install

shell

npm i @decartai/tanstack-ai-adapter

shell

pnpm add @decartai/tanstack-ai-adapter

shell

yarn add @decartai/tanstack-ai-adapter

shell

bun add @decartai/tanstack-ai-adapter

2. Set the API key

Go to the Decart Platform.
Create an account and make an API key.
Put the key in the environment of your server:

shell

DECART_API_KEY=your-api-key-here

decartImage and decartVideo read DECART_API_KEY. To pass the key in code, use createDecartImage or createDecartVideo:

import { generateImage } from "@tanstack/ai";
import { createDecartImage } from "@decartai/tanstack-ai-adapter";

const adapter = createDecartImage("lucy-pro-t2i", process.env.DECART_API_KEY!);

const result = await generateImage({
  adapter,
  prompt: "A serene mountain landscape at sunset",
  threadId: "thread-1",
});

3. Generate an image

import { generateImage } from "@tanstack/ai";
import { decartImage } from "@decartai/tanstack-ai-adapter";

const result = await generateImage({
  adapter: decartImage("lucy-pro-t2i"),
  prompt: "A futuristic cityscape at night",
  threadId: "thread-1",
});

console.log(result.images[0]?.b64Json);

You get the image as base64 data in result.images[0]?.b64Json. See Image Generation.

Video

Video generation runs as a job. You start the job, then you ask for its status until it is done.

Start the job:

import { generateVideo } from "@tanstack/ai";
import { decartVideo } from "@decartai/tanstack-ai-adapter";

const { jobId } = await generateVideo({
  adapter: decartVideo("lucy-pro-t2v"),
  prompt: "A cat playing with a ball of yarn",
  threadId: "thread-1",
});

console.log("Job started:", jobId);

Ask for the status:

import { getVideoJobStatus } from "@tanstack/ai";
import { decartVideo } from "@decartai/tanstack-ai-adapter";

const jobId = "example-job-id";
const status = await getVideoJobStatus({
  adapter: decartVideo("lucy-pro-t2v"),
  jobId,
});

console.log("Status:", status.status); // "pending" | "processing" | "completed" | "failed"

if (status.status === "completed" && status.url) {
  console.log("Video URL:", status.url);
}

The two steps together, with a status check every 5 seconds:

import { generateVideo, getVideoJobStatus } from "@tanstack/ai";
import { decartVideo } from "@decartai/tanstack-ai-adapter";

async function createVideo(prompt: string) {
  const adapter = decartVideo("lucy-pro-t2v");

  // Create the job
  const { jobId } = await generateVideo({ adapter, prompt, threadId: "thread-1" });
  console.log("Job created:", jobId);

  // Poll for completion
  let status = "pending";
  while (status !== "completed" && status !== "failed") {
    await new Promise((resolve) => setTimeout(resolve, 5000));

    const result = await getVideoJobStatus({ adapter, jobId });
    status = result.status;
    console.log(`Status: ${status}`);

    if (result.status === "failed") {
      throw new Error("Video generation failed");
    }

    if (result.status === "completed" && result.url) {
      return result.url;
    }
  }
}

const videoUrl = await createVideo("A drone shot over a tropical beach");

See Video Generation.

Models and options

Models

Model	Kind	Factory
lucy-pro-t2i	Image	decartImage, createDecartImage
lucy-pro-t2v	Video	decartVideo, createDecartVideo

Image options

Pass these in modelOptions:

import { generateImage } from "@tanstack/ai";
import { decartImage } from "@decartai/tanstack-ai-adapter";

const result = await generateImage({
  adapter: decartImage("lucy-pro-t2i"),
  prompt: "A portrait of a robot artist",
  threadId: "thread-1",
  modelOptions: {
    resolution: "720p",
    orientation: "portrait",
    seed: 42,
  },
});

Option	Type	Default	What it does
resolution	"720p"	"720p"	The output resolution.
orientation	"portrait" \| "landscape"	"landscape"	The image orientation.
seed	number	none	A seed, to get the same image again.

Video options

import { generateVideo } from "@tanstack/ai";
import { decartVideo } from "@decartai/tanstack-ai-adapter";

const { jobId } = await generateVideo({
  adapter: decartVideo("lucy-pro-t2v"),
  prompt: "A timelapse of a blooming flower",
  threadId: "thread-1",
  modelOptions: {
    resolution: "720p",
    orientation: "landscape",
    seed: 42,
  },
});

Option	Type	Default	What it does
resolution	"720p" \| "480p"	"720p"	The output resolution.
orientation	"portrait" \| "landscape"	"landscape"	The video orientation.
seed	number	none	A seed, to get the same video again.

Factories

Factory	What it does
decartImage(model, config?)	An image adapter that reads DECART_API_KEY.
createDecartImage(model, apiKey, config?)	An image adapter with an explicit API key.
decartVideo(model, config?)	A video adapter that reads DECART_API_KEY.
createDecartVideo(model, apiKey, config?)	A video adapter with an explicit API key.

Each factory takes an optional config.baseUrl, for a custom endpoint:

import { createDecartImage, type DecartImageConfig } from "@decartai/tanstack-ai-adapter";

const config: Omit<DecartImageConfig, "apiKey"> = {
  baseUrl: "https://api.decart.ai", // Optional, for custom endpoints
};

const adapter = createDecartImage("lucy-pro-t2i", process.env.DECART_API_KEY!, config);