Sogni AIInstall

Summary

Generate AI-created videos using a specified video model and detailed textual prompts, with customization options for video characteristics and workflow inputs.

Inputs

• Model Search — Optional text to filter available video models by name or tag.
• Model (required) — Select or paste the ID of the video generation model to use.
• Positive Prompt (required) — Text description of the desired video content.
• Network — Choose between 'Fast' or 'Relaxed' generation networks influencing speed, cost, and timeout defaults.
• Additional Fields — Configure detailed video generation parameters such as negative prompt, style prompt, number of videos (1-4), frames (10-120), duration in seconds (optional), frames per second (10-60), inference steps, guidance strength, motion shift, sampler and scheduler overrides, seed for reproducibility, plus workflow inputs referencing binary media properties (images, audio, video), workflow-specific controls (e.g., video/audio start offsets, trim frame, keyframe strengths, ControlNet usage and settings), output format and size options, and advanced token type selection and timeout settings.

Output shape

a binary property containing one or more generated videos in MP4 format, or URLs if downloading is disabled

The node can generate multiple videos simultaneously (up to 4) and offers an option to download videos as binary data within the workflow to prevent URL expiration. All generation parameters are passed to the Sogni AI SDK with model-aware handling. Timeouts default based on chosen network if unspecified. Reference media inputs allow complex workflows involving image/video/audio inputs for animation, interpolation, or audio-driven generation.

Links

• n8n community nodes installation guide