WaveSpeed Video Generation

WaveSpeed is a cloud API for AI video generation. Unlike RunPod containers, WaveSpeed runs on managed infrastructure — no GPU setup required. Generate videos from text prompts, images, or audio (for lip-sync).

Overview

WaveSpeed provides multiple AI video models accessible through a unified API. Plinken integrates with WaveSpeed for:

Text-to-Video — Generate video from text descriptions
Image-to-Video — Animate a reference image
Audio-to-Video (Lipsync) — Generate lip-synced talking head video from audio

Access

Open the WaveSpeed panel from Create > Video (WaveSpeed) or:

pl.cmd("create:show-wavespeed")

Setup

Sign up at wavespeed.ai
Get your API key from the WaveSpeed dashboard
Enter the API key in Plinken Settings

Commands

The WaveSpeed integration uses IPC commands for asset management:

Command	Description
`wavespeed:import-file`	Import a local file as WaveSpeed input
`wavespeed:get-assets`	List available assets (uploaded files)
`wavespeed:get-signed-url`	Get a signed URL for uploading
`wavespeed:import-video`	Import generated video to the timeline
`wavespeed:load-history`	Load generation history
`wavespeed:save-history`	Save generation history

Lua API

-- Open WaveSpeed panel
pl.cmd("create:show-wavespeed")

-- Import generated video to timeline
pl.cmd("wavespeed:import-video", {
  url = "https://...",
  name = "Generated Scene"
})

-- Get available assets
pl.cmd("wavespeed:get-assets")

Generation Workflow

Text-to-Video

Enter a text prompt describing the desired video
Select a model and set parameters (duration, resolution)
Click Generate
Preview the result
Import to timeline or save to disk

Image-to-Video

Upload or select a reference image
Optionally add a text prompt to guide the animation
Select a model and parameters
Generate and preview
Import to timeline

Audio-to-Video (Lipsync)

Select an audio clip (speech)
Provide a reference face image
The model generates a talking head video synced to the audio
Import the result to a video track

Asset Management

WaveSpeed uses a cloud asset system for input files:

Operation	Description
Upload	Send local files to WaveSpeed storage via signed URL
List	View all uploaded assets
Import	Pull generated results into your Plinken project
Register	Register a WaveSpeed asset as a Plinken project asset

History

Generation history is persisted locally, allowing you to:

Review past generations
Re-download previous results
Re-run prompts with modified parameters
Track generation costs

Importing Results

Generated videos can be:

Target	Description
Timeline	Place on a video track at the playhead
Compositor	Add as a video layer source
Asset Library	Register as a reusable project asset
Local Disk	Save the video file to a directory

Advantages Over RunPod

Feature	WaveSpeed	RunPod
Setup	API key only	Launch container, wait for startup
Cost model	Per-generation	Per-GPU-hour
Startup time	Instant	2–5 minutes
GPU management	None	Manual start/stop
Available models	Curated catalog	Any model you install
Customization	Limited to API params	Full ComfyUI flexibility

Use WaveSpeed when: You want quick, hassle-free video generation. Use RunPod when: You need custom models, workflows, or full control.

Create Overview — All AI tools
Container Manager — RunPod GPU containers
ComfyUI — Node-based AI workflows