ComfyUI Prompt-Relay is a custom node for ComfyUI designed to simplify the creation of complex prompts for AI video generation. It addresses the challenge of distributing video content across multiple segments by providing flexible syntax for defining prompt segments and automatically calculating frame allocation, benefiting users who need fine-grained control over video narrative progression.

How It Works

The core of Prompt-Relay is its "Smart node," which interprets a structured prompt to distribute video frames across defined segments. It offers two primary syntax styles: inline, using pipe (|) separators for segments, and block, employing newline headers (e.g., Scene 1:). Both styles support equal distribution or proportional distribution based on specified ranges or weights. The inline syntax uses [start-end] tags for relative positioning, while block syntax uses range headers like Scene 1-2:. An optional normalize_by_tokens toggle allows segment weights to be adjusted by their CLIP token count, allocating more frames to textually richer segments. A global_prompt field can anchor persistent video details, defaulting to the first segment's content if left empty.

Quick Start & Requirements

• Primary install: Install as a ComfyUI custom node.
• Prerequisites: ComfyUI. The word2number Python package (pip install word2number) is recommended for parsing numerical words in block headers.
• Links: Original project: https://gordonchen19.github.io/Prompt-Relay/.

Highlighted Details

• Automated frame distribution across video segments via the "Smart node."
• Dual prompt syntax options: inline (pipe-separated) and block (newline headers).
• Flexible segment weighting: equal, proportional by range, or by token count.
• global_prompt field for consistent stylistic or descriptive elements throughout the video.

Maintenance & Community

• No specific details on maintenance, contributors, or community channels were found in the provided README.

Licensing & Compatibility

• No license information is specified in the README.

Limitations & Caveats

• The project is explicitly marked as "WORK IN PROGRESS," indicating potential instability or incomplete features.