Nodes/Runware.ai ComfyUI Inference API Integration/Runware Accelerator Options
ComfyUI Node

Runware Accelerator Options

A ComfyUI node in Runware/Params with 21 inputs and 1 output.

By Runware·Created 2 years ago·Updated about a month ago· 140
Runware Accelerator Options
    • acceleratorOptions
    ◄cacheStartStepfalse►
    ◄cacheStartStep_value0►
    ◄cacheStartStepPercentagefalse►
    ◄cacheStartStepPercentage_value0►
    ◄cacheEndStepfalse►
    ◄cacheEndStep_value1►
    ◄cacheEndStepPercentagefalse►
    ◄cacheEndStepPercentage_value1►
    ◄cacheMaxConsecutiveSteps3►
    ◄teaCachefalse►
    ◄teaCacheDistance0.50►
    ◄dbCachefalse►
    ◄dbCacheThreshold0.25►
    ◄dbCacheSkipInterval5►
    ◄fbCachefalse►
    ◄fbCacheThreshold0.25►
    ◄deepCachefalse►
    ◄deepCacheIntervalfalse►
    ◄deepCacheInterval_value1►
    ◄deepCacheBranchIdfalse►
    ◄deepCacheBranchId_value0►
    CategoryRunware/Params

    Inputs (21)

    NameTypeDefaultDescription
    cacheStartStepoptBOOLEANfalseEnable to set cacheStartStep. Off uses the model's default.
    cacheStartStep_valueoptINT0Absolute step number to start caching. Must be less than `cacheEndStep`.
    cacheStartStepPercentageoptBOOLEANfalseEnable to set cacheStartStepPercentage. Off uses the model's default.
    cacheStartStepPercentage_valueoptINT00–99Percentage of steps to start caching. Alternative to `cacheStartStep`. Must be less than `cacheEndStepPercentage`.
    cacheEndStepoptBOOLEANfalseEnable to set cacheEndStep. Off uses the model's default.
    cacheEndStep_valueoptINT1Absolute step number to end caching. Must be greater than `cacheStartStep` and less than or equal to `steps`.
    cacheEndStepPercentageoptBOOLEANfalseEnable to set cacheEndStepPercentage. Off uses the model's default.
    cacheEndStepPercentage_valueoptINT11–100Percentage of steps to end caching. Alternative to `cacheEndStep`. Must be greater than `cacheStartStepPercentage`.
    cacheMaxConsecutiveStepsoptINT31–5Limits the maximum number of consecutive steps that can use cached computations before forcing a fresh computation.
    teaCacheoptBOOLEANfalseTeaCache acceleration for transformer-based models. Estimates step differences to skip redundant computations.
    teaCacheDistanceoptFLOAT0.500–1Controls the aggressiveness of the TeaCache feature. Lower values prioritize quality, higher values prioritize speed.
    dbCacheoptBOOLEANfalseDB Cache (CacheDiT) acceleration. Caches and reuses intermediate transformer block outputs to skip redundant computations.
    dbCacheThresholdoptFLOAT0.250–1Controls the sensitivity threshold for DB Cache. Lower values reuse cached blocks more aggressively, higher values prioritize quality.
    dbCacheSkipIntervaloptINT5Controls how many steps to skip between cache refreshes. Higher values skip more steps for faster generation at the cost of quality.
    fbCacheoptBOOLEANfalseFirst Block Cache (FBCache) acceleration. Reuses feature block computations across steps.
    fbCacheThresholdoptFLOAT0.250–1Controls the sensitivity threshold for determining when to reuse cached computations. Lower values reuse more aggressively.
    deepCacheoptBOOLEANfalseDeepCache acceleration. Skips transformer computations in certain steps to speed up generation.
    deepCacheIntervaloptBOOLEANfalseEnable to set deepCacheInterval. Off uses the model's default.
    deepCacheInterval_valueoptINT1Interval for DeepCache acceleration. A value of 2 skips every other step, 3 skips two out of three, etc.
    deepCacheBranchIdoptBOOLEANfalseEnable to set deepCacheBranchId. Off uses the model's default.
    deepCacheBranchId_valueoptINT0Branch ID for DeepCache acceleration. Determines which U-Net layers are skipped.

    Outputs (1)

    NameTypeDescription
    acceleratorOptionsRUNWARE_ACCELERATOROPTIONS—