runway-video

n8n community node for Runway's text_to_video API — model selection (Gen-4.5, Veo 3/3.1, Seedance 2, Happyhorse, Gemini Omni Flash), model-aware ratio/duration, and Seedance 2 reference video/image/audio consistency.

Documentation

n8n-nodes-runway-video

An n8n community node for Runway's video generation API, supporting multiple underlying models beyond Runway's own — including Google Veo 3/3.1, Seedance 2, Happyhorse, and Gemini Omni Flash.

Installation

Follow the n8n community nodes installation guide, or install manually:

npm install n8n-nodes-runway-video

Credentials

Requires a Runway API credential:

Field Description
API Secret Your Runway API key
Base URL Defaults to https://api.dev.runwayml.com
API Version Defaults to 2024-11-06

Operations

The Runway Text to Video node supports:

  • Create & Wait for Video — submits the generation task and polls until it succeeds or fails, returning the final output URL(s). Simplest option for most workflows.
  • Create Task Only — submits and returns immediately with a task ID.
  • Get Task Status — checks progress on a previously submitted task.
  • Cancel / Delete Task — cancels a running or queued task.

Supported models

Gen-4.5 (Runway, highest fidelity), Veo 3, Veo 3.1, Veo 3.1 Fast, Happyhorse 1.0, Seedance 2.0, Seedance 2.0 Fast, Seedance 2.0 Mini, and Gemini Omni Flash.

Fields adapt to the selected model:

  • Negative Prompt is only used by the Veo 3 family.
  • Reference video/image/audio consistency (for character or style continuity across generations) is only supported by the Seedance 2 family.
  • There's no separate "language" field — prompt text is written in whatever language you want reflected in the video, including spoken audio for audio-capable models.

License

MIT

Discussion