One SDK. Every AI media model.

Ship AI-generated video,
image & voice — in 3 lines of code

One SDK for Sora, Veo, Kling, DALL·E, Flux, ElevenLabs & 25+ more models. Stop juggling AI APIs. Start shipping.

npm GitHub Docs

Get Started

$npm install zyka-sdk

generate.ts

const client = new ZykaClient({ apiKey: process.env.ZYKA_API_KEY });

const video = await client.createVideo({ 
  model: 'sora',
  prompt: 'a cat surfing a wave'
 });

// video.url — done.

Every model, one interface

30+ models, one interface

Video

SoraVeo 2KlingWANHailuo

Image

DALL·E 3FluxStable DiffusionNano BananaIdeogram

Voice

ElevenLabsChatterboxMiniMaxPlayHT

DX that gets out of your way

The details you shouldn't have to think about — handled.

Auto-Wait

await = done. No polling, no webhooks, no status checks.

const video = await client.createVideo({
  model: 'kling',
  prompt: 'sunset timelapse'
});
// Resolves when generation completes

Local Files

Pass local file paths directly. We handle the upload.

const result = await client.createImage({
  model: 'flux',
  image: './photo.png',
  prompt: 'make it watercolor'
});

Auto-Download

Output straight to disk. No manual fetch needed.

await client.createVideo({
  model: 'wan',
  prompt: 'drone flyover',
  output: './result.mp4'
});

CLI Built-in

Generate from your terminal. Script it, pipe it, cron it.

$ npx zyka generate video \
    -m wan \
    -p "sunset over mountains" \
    -o result.mp4

Same pattern, every media type

Learn it once. Use it everywhere.

VideoImageVoiceApps

video.ts

1import { ZykaClient } from 'zyka-sdk'
2
3const client = new ZykaClient()
4
5const video = await client.createVideo({
6  model: 'sora',
7  prompt: 'cinematic drone shot of a coral reef'
8})
9
10console.log(video.url)

image.ts

1import { ZykaClient } from 'zyka-sdk'
2
3const client = new ZykaClient()
4
5const image = await client.createImage({
6  model: 'flux',
7  prompt: 'minimalist logo, white on black',
8  output: './logo.png'
9})

voice.ts

1import { ZykaClient } from 'zyka-sdk'
2
3const client = new ZykaClient()
4
5const audio = await client.createTTS({
6  model: 'elevenlabs',
7  text: 'Welcome to the future of AI.',
8  voice: 'aria'
9})

apps.ts

1import { ZykaClient } from 'zyka-sdk'
2
3const client = new ZykaClient()
4
5const result = await client.run({
6  app: 'face-swap',
7  source: './source.mp4',
8  target: './face.jpg',
9  output: './swapped.mp4'
10})

Not just generation — a full AI media toolkit

Pre-built AI apps you can call with a single function.

Face Swap

Swap faces in video or images

Virtual Try-On

Try clothing on any person

Video Upscaler

Upscale resolution with AI

Caption Generator

Auto-generate subtitles

Video Dubbing

Translate & dub any video

B-Roll Insertion

Generate contextual B-roll

Background Remover

Remove or replace backgrounds

Image Restore

Enhance & restore old photos

How it works

Install the SDK

$ npm install zyka-sdk

Set your API key

$ export ZYKA_API_KEY=sk_live_...

Generate anything

$ client.createVideo() / createImage() / createTTS()

Raw APIs vs Zyka SDK

Less boilerplate. More shipping.

Feature	Raw APIs	Zyka SDK
Models supported	1 per integration	30+ unified
Polling for results	Manual	Handled
File upload handling	Manual	Handled
Download results to disk	Manual	Handled
Auth per provider	Manual	Handled
CLI access	Handled	Built-in

Start shipping in minutes

Get your API key and generate your first video, image, or voice clip in under 60 seconds.