Prompt and seed travel — v1.0

Prompt and seed travel — v1.0 is a ComfyUI workflow for generating images across changing prompt and seed points; it takes positive and negative prompts, timing curves, per-point seeds, image count, video framerate, CFG controls, and a selected model, and produces generated image

Choose among eight timing curves and four time-compression settings, from off to pow(5).

Set a seed for each point, then choose the image count used for generation.

The output area places video at the top, generated images at the bottom, and an output-density graph in the middle.

The described setup uses an image count of 300, which controls how many images are generated.

The outputs include the generated image set, a video, and a graph showing output density.

This workflow is v1.0, published by MkDaniel, and lists ZImageBase as its base model.

Before use, verify that these required model files are available: qwen_3_4b.safetensors, ae.safetensors, z_image_bf16.safetensors, and fairy_forest.safetensors.

Suggestion · not verified

It requires two node packs: comfyui-mxtoolkit and more_math.

Some models require CFG and negative prompts, while others can work without them; toggle those controls for the selected model.

Suggestion · not verified

You can exchange nodes for models requiring more CLIP models, or use a checkpoint loader for SDXL or similar models.

Leave the Adaptive module off to start; enabling it requires at least 41 images, and use with fewer than 60 images was not tested.

Suggestion · not verified

Be aware that some timing curves do not work with noise and can produce grey images.

Suggestion · not verified

The output area can become laggy after generation, and the density graph may not exactly reflect output when Adaptive stages are off.

Suggestion · not verified