ComfyUI > Nodes > ComfyUI-WanVideoWrapper > WanVideo EasyCache

ComfyUI Node: WanVideo EasyCache

Class Name

WanVideoEasyCache

Category
WanVideoWrapper
Author
kijai (Account age: 2871days)
Extension
ComfyUI-WanVideoWrapper
Latest Updated
2026-05-05
Github Stars
6.41K

How to Install ComfyUI-WanVideoWrapper

Install this extension via the ComfyUI Manager by searching for ComfyUI-WanVideoWrapper
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter ComfyUI-WanVideoWrapper in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

WanVideo EasyCache Description

Optimizes video processing models by caching diffusion model outputs for faster inference in WanVideoWrapper suite.

WanVideo EasyCache:

WanVideoEasyCache is a node designed to optimize the performance of video processing models by caching the outputs of diffusion models. This caching mechanism is particularly beneficial in scenarios where repeated computations can be avoided, thereby speeding up the inference process. The node is part of the WanVideoWrapper suite and leverages the EasyCache method to manage caching efficiently. By setting specific thresholds and steps, you can control when and how the caching is applied, ensuring that the model's performance is enhanced without compromising the quality of the output. The primary goal of WanVideoEasyCache is to reduce computational overhead and improve processing times, making it an essential tool for AI artists working with complex video models.

WanVideo EasyCache Input Parameters:

easycache_thresh

This parameter determines the strength of the caching mechanism applied to the diffusion model's output. It is a floating-point value that must be non-negative, with a default of 0.015. The minimum value is 0.0, and the maximum is 1.0, with increments of 0.001. A higher threshold means more aggressive caching, which can speed up processing but may affect the output quality if set too high.

start_step

The start_step parameter specifies the step at which the EasyCache begins to apply caching. It is an integer value with a default of 10, a minimum of 0, and a maximum of 9999. This allows you to delay the caching process until a certain point in the model's execution, which can be useful for ensuring that early steps, which are often crucial for establishing the model's baseline, are not skipped.

end_step

This parameter defines the step at which the caching process should end. It is an integer with a default value of -1, which indicates that caching should continue until the end of the process. The minimum value is -1, and the maximum is 9999. Setting a specific end step can help manage resources and ensure that caching is only applied when beneficial.

cache_device

The cache_device parameter allows you to choose the device where the cache will be stored. The options are "main_device" and "offload_device," with "offload_device" as the default. This choice can impact the speed and efficiency of the caching process, as different devices may have varying capabilities and resource availability.

WanVideo EasyCache Output Parameters:

cache_args

The cache_args output parameter is a dictionary containing the configuration settings for the caching process. It includes details such as the cache type, threshold, start and end steps, and the device used for caching. This output is crucial for ensuring that the caching process is executed according to the specified parameters, allowing for optimized performance and resource management.

WanVideo EasyCache Usage Tips:

  • Adjust the easycache_thresh parameter carefully to balance between performance and output quality. A lower threshold may result in less aggressive caching, preserving more detail in the output.
  • Use the start_step and end_step parameters to fine-tune when caching should occur. Starting caching too early might skip important initial steps, while ending it too late might not provide the desired performance boost.
  • Consider the capabilities of your hardware when selecting the cache_device. Offloading to a secondary device can free up resources on the main device, but ensure that the offload device is capable of handling the cache efficiently.

WanVideo EasyCache Common Errors and Solutions:

"Invalid cache device selected"

  • Explanation: This error occurs when an unsupported device is chosen for caching.
  • Solution: Ensure that the cache_device parameter is set to either "main_device" or "offload_device".

"Threshold value out of range"

  • Explanation: The easycache_thresh value is set outside the allowed range.
  • Solution: Adjust the easycache_thresh to be within the range of 0.0 to 1.0.

"Start step greater than end step"

  • Explanation: The start_step is set to a value greater than end_step, causing a logical error in the caching process.
  • Solution: Ensure that start_step is less than or equal to end_step to maintain a valid caching sequence.

WanVideo EasyCache Related Nodes

Go back to the extension to check out more related nodes.
ComfyUI-WanVideoWrapper
RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

WanVideo EasyCache