Quickstart

Get up and running with Synthome SDK in just a few minutes.

Installation

Install the SDK using your preferred package manager:

npm install @synthome/sdk

yarn add @synthome/sdk

pnpm add @synthome/sdk

bun add @synthome/sdk

Setup API Keys

1. Synthome API Key

Get your Synthome API key from the dashboard:

export SYNTHOME_API_KEY="your-synthome-api-key"

2. Provider API Keys

Synthome orchestrates AI models across multiple providers. You need API keys for the providers you want to use.

Option A: Dashboard (Recommended)

Add your provider API keys in the Synthome dashboard. This keeps your keys secure and managed in one place.

Option B: Environment Variables

# For Replicate models
export REPLICATE_API_KEY="your-replicate-key"

# For Fal models
export FAL_KEY="your-fal-key"

# For ElevenLabs audio
export ELEVENLABS_API_KEY="your-elevenlabs-key"

# For Hume TTS
export HUME_API_KEY="your-hume-key"

Option C: Pass in Code

const model = videoModel("bytedance/seedance-1-pro", {
  provider: "replicate",
  apiKey: "your-replicate-key",
});

Priority Order: Model-level API key → Dashboard keys → Environment variables

Your First Pipeline

Let's generate a simple video:

import { compose, generateVideo, videoModel } from "@synthome/sdk";

const execution = await compose(
  generateVideo({
    model: videoModel("bytedance/seedance-1-pro", "replicate"),
    prompt: "A serene mountain landscape at sunrise, cinematic",
    duration: 5,
  }),
).execute();

console.log("Video URL:", execution.result?.url);

Generate an Image

import { compose, generateImage, imageModel } from "@synthome/sdk";

const execution = await compose(
  generateImage({
    model: imageModel("google/nano-banana", "fal"),
    prompt: "A futuristic city skyline at night",
  }),
).execute();

console.log("Image URL:", execution.result?.url);

Generate Audio (Text-to-Speech)

import { compose, generateAudio, audioModel } from "@synthome/sdk";

const execution = await compose(
  generateAudio({
    model: audioModel("elevenlabs/turbo-v2.5", "elevenlabs"),
    text: "Welcome to Synthome, the composable AI media toolkit.",
    voiceId: "EXAVITQu4vr4xnSDxMaL", // Sarah voice
  }),
).execute();

console.log("Audio URL:", execution.result?.url);

Combine Multiple Operations

The real power of Synthome is composing multiple operations. You can mix direct URLs with generated media:

import { compose, generateVideo, merge, videoModel } from "@synthome/sdk";

const execution = await compose(
  merge([
    // Use an existing video URL
    "https://example.com/intro.mp4",
    // Generate new videos
    generateVideo({
      model: videoModel("bytedance/seedance-1-pro", "replicate"),
      prompt: "Scene 1: A rocket launching into space",
    }),
    generateVideo({
      model: videoModel("bytedance/seedance-1-pro", "replicate"),
      prompt: "Scene 2: Earth from orbit",
    }),
  ]),
).execute();

console.log("Merged video:", execution.result?.url);

Generated videos run in parallel, and the merge waits for all inputs before combining them.

Use Webhooks (Async)

For long-running pipelines, use webhooks instead of polling:

const execution = await compose(
  generateVideo({
    model: videoModel("bytedance/seedance-1-pro", "replicate"),
    prompt: "An epic cinematic scene",
  }),
).execute({
  webhook: "https://your-server.com/webhook",
  webhookSecret: "your-secret", // Optional: for signature verification
});

// Returns immediately with execution ID
console.log("Execution ID:", execution.id);