Very Good FFmpeg
How it worksPricingDocsBlog
Documentation
API Reference
Documentation
Documentation
Getting Started
Fundamentals
Advanced Topics
Temporary FilesCommand ChainingHardware SelectionWebhooksFile Reference Styles
Integrations
Advanced Topics
  1. Advanced Topics
  2. Hardware Selection

Hardware Selection

Choose CPU or NVIDIA GPU workers for your jobs.

Choose the underlying compute for your job using the machine property.

Options

  • cpu (default): Dedicated 16-core instances with high-clock-speed CPUs. Best for most encoding tasks.
  • nvidia: Instances equipped with an NVIDIA L4 GPU. Use this when your FFmpeg command utilizes GPU encoders like h264_nvenc or hevc_nvenc.

CPU Example

Default CPU job
{
  "input_files": { "input.mp4": "https://example.com/video.mp4" },
  "output_files": ["output.mp4"],
  "ffmpeg_commands": ["-i inputs/input.mp4 -c:v libx264 outputs/output.mp4"]
}

NVIDIA Example

GPU-accelerated encode
{
  "input_files": { "input.mp4": "https://example.com/video.mp4" },
  "output_files": ["output.mp4"],
  "machine": "nvidia",
  "ffmpeg_commands": ["-i inputs/input.mp4 -c:v h264_nvenc outputs/output.mp4"]
}

Notes

GPU workers only help when your FFmpeg command uses GPU decode, encode, or filters. Setting "machine": "nvidia" does not make normal CPU filters run on the GPU.

Temporary Files

Upload local media files.

Command Chaining

Run multi-step pipelines on the same machine.

Command Chaining

Run multiple FFmpeg commands sequentially on the same machine.

Webhooks

Get notified when your jobs complete.

On this page

OptionsCPU ExampleNVIDIA ExampleNotes