MachGen logo

Documentation

Build with MachGen

The fastest AI video inference platform. Access world-class models through a simple API — or deploy an optimized inference stack for your own models.

5 production models
REST API + Python SDK
Pay-as-you-go credits

Get started in 3 steps

From zero to your first generated video in under 5 minutes.

01
Install the SDK
pip install machgen
02
Set your API key
export MACHGEN_API_KEY=your_api_key
03
Generate a video
import machgen

result = machgen.run(
  model="wan-2.2",
  prompt="A red panda exploring a misty forest at dawn",
)
print(result.video_url)

Explore the docs

Everything you need to integrate, deploy, and scale.

Supported models

All models are optimized for maximum throughput and minimum latency on MachGen infrastructure.

Wan 2.2Popular

High-fidelity text-to-video and image-to-video with exceptional motion quality.

Text-to-VideoImage-to-Video
Hunyuan 1.5High Quality

Tencent's flagship video model delivering cinematic quality at scale.

Text-to-Video720p
LTX 2.3Fast

Optimized for speed without sacrificing output quality. Ideal for real-time workflows.

Text-to-VideoLow Latency
KlingCinematic

Exceptional at generating smooth, realistic motion with strong prompt adherence.

Text-to-VideoImage-to-Video
ViduCreative

Best-in-class for creative and stylized video generation with rich visual detail.

Text-to-VideoStylized

Need help?

Can't find what you're looking for? Our technical team is here to help.