​
Installation

npm install modelslab

​
Basic setup / Connecting to Client

Create client to connect to ModelsLab API. 
// Method 1: Direct API key
import { Client } from "modelslab";
const client = new Client("your-api-key");

// Method 2: Environment variable
// Client reads from process.env.API_KEY when called without args
const clientFromEnv = new Client();

// Method 3: With custom settings (retries, timeout)
const clientCustom = new Client("your-api-key", 5, 10); // 5 retries, 10s timeout
You can access client properties: 
console.log(client.baseUrl);      
console.log(client.fetchRetry);   
console.log(client.fetchTimeout);

​
Select API Client

import { Community, Audio, Video, DeepFake, ImageEditing } from "modelslab";

// Create instances
const community = new Community(client.key);
const audio = new Audio(client.key);
const video = new Video(client.key);
const deepfake = new DeepFake(client.key);
const imageEditing = new ImageEditing(client.key);
For enterprise/dedicated endpoints: 
// Enable enterprise mode
const enterpriseCommunity = new Community(client.key, true);
// Uses enterprise endpoints, e.g. https://modelslab.com/api/v1/enterprise/images/

​
Quick Start — Your First AI Image

import { Client, Community } from "modelslab";

const client = new Client("your-api-key-here");
const community = new Community(client.key);

const result = await community.textToImage({
  key: client.key,
  prompt: "A beautiful sunset over mountains",
  model_id: "flux",
  width: 512,
  height: 512,
  samples: 1,
});

console.log("Generated image:", result.output[0]);

​
Available APIs (summary)

  • Community — community models (text→image, image→image, controlnet, inpainting)
  • Audio — text-to-audio, text-to-speech, voice2voice, voice cover, music generation, lyrics, song generator, speech-to-text, SFX
  • Video — text-to-video, image-to-video
  • DeepFake — face swap and video swap endpoints
  • ImageEditing — outpainting, background removal, super resolution, fashion, object removal, facegen, inpainting, headshot, flux headshot

​
Community API Usage

import { Community } from "modelslab";

const community = new Community(client.key);

// Text to Image
const image = await community.textToImage({
  key: client.key,
  prompt: "A futuristic city at night",
  model_id: "stable-diffusion-v1-5",
  width: 512,
  height: 512,
  samples: 1,
  num_inference_steps: 20,
  guidance_scale: 7.5,
});

// Image to Image
const imageToImage = await community.imageToImage({
  key: client.key,
  prompt: "A painting in Van Gogh style",
  init_image: "base64-encoded-image",
  model_id: "stable-diffusion-v1-5",
  strength: 0.8,
});

// Inpainting (Fill masked areas)
const inpainting = await community.inpainting({
  key: client.key,
  prompt: "A red car",
  init_image: "base64-encoded-image",
  mask_image: "base64-encoded-mask",
  model_id: "stable-diffusion-v1-5",
});

// ControlNet (Guided generation)
const controlnet = await community.controlnet({
  key: client.key,
  prompt: "A realistic portrait",
  controlnet_model: "canny",
  controlnet_conditioning_scale: 1.0,
  model_id: "stable-diffusion-v1-5",
});

​
Audio API Usage

import { Audio } from "modelslab";

const audio = new Audio(client.key);

// Text to Audio
const t2a = await audio.textToAudio({...});

// Text to Speech
const tts = await audio.textToSpeech({...});

// Voice to Voice
const v2v = await audio.voice2voice({...});

// Voice Cover
const vc = await audio.voiceCover({...});

// Music Generation
const mg = await audio.musicGen({...});

// Lyrics generation
const lyrics = await audio.lyricsGen({...});

// Song generation
const song = await audio.songGenerator({...});

// Speech to Text
const stt = await audio.speechToText({...});

// SFX generation
const sfx = await audio.sfxGen({...});

​
Video API Usage

import { Video } from "modelslab";

const video = new Video(client.key);

// Text to Video
const t2v = await video.textToVideo({...});

// Image to Video
const i2v = await video.imageToVideo({...});

​
DeepFake API Usage

import { DeepFake } from "modelslab";

const deepfake = new DeepFake(client.key);

// Specific Face Swap
const sfs = await deepfake.specificFaceSwap({...});

// Multiple Face Swap
const mfs = await deepfake.multipleFaceSwap({...});

// Specific Video Swap
const svs = await deepfake.specificVideoSwap({...});

// Single Video Swap
const single = await deepfake.singleVideoSwap({...});

​
Image Editing API Usage

import { ImageEditing } from "modelslab";

const imageEditing = new ImageEditing(client.key);

// Outpainting
const out = await imageEditing.outpainting({...});

// Background Remover
const br = await imageEditing.backgroundRemover({...});

// Super Resolution
const sr = await imageEditing.superResolution({...});

// Fashion
const fashion = await imageEditing.fashion({...});

// Object Remover
const obj = await imageEditing.objectRemover({...});

// Facegen
const facegen = await imageEditing.facegen({...});

// Inpainting
const inp = await imageEditing.inpainting({...});

// Headshot
const headshot = await imageEditing.headshot({...});

// Flux Headshot
const flux = await imageEditing.fluxHeadshot({...});