Wan 2.2 · Hunyuan Video · LTX · Mochi — one API

Generate cinematic AI video in minutes

Prompt-to-video and face swap, powered by the best open models and a queue that scales from one GPU to a fleet.

Everything you need to ship AI video

Cinematic video generation

Turn a single prompt into fluid, coherent video with state-of-the-art open models.

Face swap pipelines

Swap faces across images and video with production-grade identity preservation.

Interchangeable models

Wan, Hunyuan, LTX, Mochi — pick per job. New models plug in without code changes.

Live progress

Watch every denoising step stream to your dashboard in real time over SSE.

Scales with your GPUs

Stateless workers pull from a distributed queue. Add containers, get throughput.

Built for teams

API keys, webhooks, credits and billing-ready architecture from day one.

Pick the right model for every shot

2 credits / sec

AnimateDiff Lightning

Ultra-fast 4-step motion model that runs on 6GB GPUs

10 credits / sec

Wan 2.2

High-fidelity text-to-video with strong motion coherence

14 credits / sec

Hunyuan Video

Cinematic quality with excellent prompt adherence

6 credits / sec

LTX Video

Fast generation optimized for quick iterations

12 credits / sec

Mochi 1

Smooth photorealistic motion at 30fps

Start with 100 free credits

No credit card required. Your first generations are on us.