Documentation

One API for 255+ generative models

Image, video, audio and text through one consistent surface. Call any model from the TypeScript or Python SDK, the REST API, the gen-ai CLI, or an MCP-connected agent.

Image: 78
Video: 102
Audio: 32
Text: 43

Start here#

QuickstartCreate a client and run your first image, text and video generation.Get an API keyCreate a key in the console and pass it to the SDK, the API or the CLI.Explore modelsEvery model with its parameters, rates and copy-paste examples.

Hello, world#

Every model runs through the same generate() call — only the model id and its parameters change.

import { createClient } from '@picsart/ai-sdk';

const ai = createClient({
  apiKey: process.env.PICSART_API_KEY,
  apiUrl: 'https://api.picsart.com',
});

const result = await ai.generate('flux-2-pro', {
  prompt: 'A serene mountain lake at golden hour',
});

console.log(result.url);

Guides#

Image generationText-to-image, editing, and reference images.Video generationText-to-video and image-to-video, with progress tracking.Audio generationSpeech, music, and sound effects.Text & analysisLLM text generation and vision.

Interfaces#

Reach the same models through whichever surface fits how you work.

SDKTypeScript and Python. ai.generate() — the recommended path.REST APICall the workflows over HTTP from any language.CLIThe gen-ai command for scripting and automation.MCPGenerate from Claude Code, Cursor, and other agents.

Concepts#

Files & DriveInput URLs, the CDN, and auto-saving to Drive.