The code-native video gateway for developers and agents

Route prompts to rendering engines and orchestration models through one OpenAI-compatible API

Featured models

Experiment with leading LLMs for video orchestration that aren't available via your existing subscription.

Claude Opus 4.8
Gemini 3.5 Flash
MiniMax M3
Kimi K2.6
DeepSeek V4 Pro
+9 more

One API for code-driven video generation

Define video in code—HTML with HyperFrames, React with Remotion. The output is linked to an editable artifact you can tweak and render, unlike diffusion-based video models that can't be modified with such precision. .

Provider Routing

Route to multiple providers with automatic failover. When one provider is down, requests route to the next available option.

Token-Based Billing

Pay for orchestration tokens, not GPU hours. Transparent pricing per model with no subscriptions or minimum commitments.

OpenAI-Compatible

Drop-in replacement for the OpenAI SDK. Change the base URL and start generating videos with your existing code.

Webhook Delivery

Get notified when videos complete. Pass webhook_url per request, with retries and signature verification.

Start generating in three steps

Get an API key, load credits, and send your first generation request.

1

Create an account

Sign up to get an API key for your app or agent. No credit card required to create an account.

Create account
2

Add credits

Purchase credits in your dashboard when you are ready to generate. Credits apply to any orchestration model or rendering engine.

Add credits
3

Make a request

Send a prompt with your chosen model and engine. Pipevideo handles provider routing, rendering, and webhook delivery when the video is ready.

Read API docs

Why Pipevideo?

Creating videos with Claude, Cursor, or Claude Code is not scalable once you leave your machine.

Pipevideo providers handle the hard parts of video generation, like server rendering and agent harnesses, so you can focus on your core use case.

Welcome to code-native video's text-to-video moment.

Video Generator Agent

Install and use Pipevideo with HyperFrames and your chosen LLM.

import { Pipevideo } from '@pipevideo/sdk';
import { NextResponse } from 'next/server';

const client = new Pipevideo({
  apiKey: process.env.PIPEVIDEO_API_KEY
});

export async function POST(req: Request) {
  const { prompt } = await req.json();

  const video = await client.responses.create({
    model: 'moonshotai/kimi-k2.5',
    engine: 'hyperframes',
    input: prompt,
    provider: { order: ['pipevideo'] },
  });

  return NextResponse.json(video);
}

Recent posts

Product updates, guides, and insights from the Pipevideo team.

Jun9

Unified Video Generation API for Developers

How Pipevideo simplifies video generation by providing a single OpenAI-compatible API for multiple video engines and orchestration models.

Read more
Jun8

Routing Prompts to Video Engines with Orchestration Models

Learn how Pipevideo routes your prompts to the right video generation engine and orchestration model for optimal results.

Read more
Jun7

Comparing Video Generation Latency and Cost Across Providers

How we built the Pipevideo Rankings dashboard to provide real-time insights into video generation performance and pricing.

Read more

Ready to generate?

Get your API key and start building with code-driven video.

Video Gateway for Developers | Pipevideo | Pipevideo