Extensions/ComfyUI-latent-ops
ComfyUI Extension

ComfyUI-latent-ops

A collection of nodes for manipulating LATENT in ComfyUI.

By hnmr293·Created about a year ago·Updated about a year ago· 2
hnmr293/ComfyUI-latent-ops
Nodes46
On cloudLocal install
Categoryhnmr/latent_ops
Stars2
Updatedabout a year ago

Nodes (46)

AssertDims

A tripwire for latent shapes that fails loudly instead of weirdly

hnmr/latent_ops
AssertShape

Pin the exact latent shape and stop guessing why downstream nodes break

hnmr/latent_ops
GetSigma

The invisible number that wrecks ELLA prompts

hnmr/latent_ops
Latent01ToImage

Peek inside a 0-1 latent without paying for a VAE decode

hnmr/latent_ops
Latent11ToImage

The visualizer for -1..1 latents (it scales, the 01 node doesn't)

hnmr/latent_ops
LatentOperationAdd

Add two latents together — but only when something applies the op

hnmr/latent_ops
LatentOperationAddBroadcast

Nudge every latent value by a constant

hnmr/latent_ops
LatentOperationApplyCFG

Re-apply guidance to an already-sampled latent

hnmr/latent_ops
LatentOperationCELU

The ELU variant that stays smooth below zero

hnmr/latent_ops
LatentOperationClamp

Keep latent values inside a range

hnmr/latent_ops
LatentOperationClampMax

Clip the ceiling on your latent

hnmr/latent_ops
LatentOperationClampMin

Put a floor under your latent

hnmr/latent_ops
LatentOperationCosh

The hyperbolic curve you'll reach for exactly once

hnmr/latent_ops
LatentOperationELU

The activation that keeps negatives alive

hnmr/latent_ops
LatentOperationFill

Reset a latent to a constant

hnmr/latent_ops
LatentOperationGELU

The activation your diffusion model already loves

hnmr/latent_ops
LatentOperationHardSigmoid

Sigmoid's fast, piecewise cousin

hnmr/latent_ops
LatentOperationHardSwish

The cheap, smooth-sounding Swish

hnmr/latent_ops
LatentOperationHardTanh

Tanh with training wheels, and a -1..1 guarantee

hnmr/latent_ops
LatentOperationInstanceNorm

The zero-input node that normalizes style away

hnmr/latent_ops
LatentOperationInterpolate

Blend two latents, the way morphing was always meant to work

hnmr/latent_ops
LatentOperationLayerNorm

When your latents need a recalibration, not a rescale

hnmr/latent_ops
LatentOperationLeakyReLU

The ReLU that doesn't murder your negatives

hnmr/latent_ops
LatentOperationLogistic

It's log-sigmoid, not the sigmoid you expect

hnmr/latent_ops
LatentOperationMish

The smooth, self-regularizing activation for latents

hnmr/latent_ops
LatentOperationMul

Multiply one latent by another latent, element by element

hnmr/latent_ops
LatentOperationMulBroadcast

The humble scalar multiply that does most of the work

hnmr/latent_ops
LatentOperationNormalize

One global z-score to pull a latent back to earth

hnmr/latent_ops
LatentOperationNormalizeAlongAxis

Normalize the channel, not the whole world

hnmr/latent_ops
LatentOperationNormalizeMinMax

Squash your latent into 0–1 and hope nothing is an outlier

hnmr/latent_ops
LatentOperationNormalizePercentile

Min-max normalization that ignores the weird values

hnmr/latent_ops
LatentOperationReLU

The nuclear option for latents, and why you should think twice

hnmr/latent_ops
LatentOperationReLU6

ReLU with a cap that your latent will almost never hit

hnmr/latent_ops
LatentOperationReshape

How to break a latent's shape (and maybe your workflow)

hnmr/latent_ops
LatentOperationRoll

Shift your latent around and wrap it back like a scrolling marquee

hnmr/latent_ops
LatentOperationSELU

The self-normalizing activation that mostly normalizes for you

hnmr/latent_ops
LatentOperationSigmoid

The real S-curve, for squashing latents into 0–1

hnmr/latent_ops
LatentOperationSiLU

The modern neural-net activation, applied to your latent

hnmr/latent_ops
LatentOperationSinh

The steepest, most aggressive nonlinearity in the box

hnmr/latent_ops
LatentOperationSlice

Carve a chunk out of your latent — and learn the '0 means end' quirk

hnmr/latent_ops
LatentOperationSoftmax

Turn a latent axis into probabilities, and wonder why

hnmr/latent_ops
LatentOperationSoftmin

Softmin is the softmax that cheers for the smallest value

hnmr/latent_ops
LatentOperationSoftplus

The smooth ReLU that never quite zeroes out

hnmr/latent_ops
LatentOperationSoftsign

The tanh that takes its time at the edges

hnmr/latent_ops
LatentOperationSplitCFG

Pry the cond and uncond apart when one latent is secretly two

hnmr/latent_ops
LatentOperationTanh

Tanh your latent before it does something dramatic

hnmr/latent_ops
Readme

ComfyUI-latent-ops

日本語 ver.

A collection of nodes for manipulating LATENT in ComfyUI.

All nodes are added under hnmr/latent_ops.

Nodes

| Name | Function | | --- | --- | | AssertDims | Verify that the input LATENT has the specified number of dimensions | | AssertShape | Verify that the input LATENT has the specified shape | | LatentOperationNormalizeAlongAxis | Normalize along a specific axis | | LatentOperationNormalize | Normalize across the entire tensor | | LatentOperationLayerNorm | Apply layer normalization | | LatentOperationInstanceNorm | Apply instance normalization | | LatentOperationNormalizeMinMax | Normalize between minimum and maximum values | | LatentOperationNormalizePercentile | Normalize based on percentiles | | LatentOperationSigmoid | Apply sigmoid function | | LatentOperationHardSigmoid | Apply hard sigmoid function | | LatentOperationLogistic | Apply logistic function | | LatentOperationTanh | Apply tanh function | | LatentOperationHardTanh | Apply hard tanh function | | LatentOperationSinh | Apply sinh function | | LatentOperationCosh | Apply cosh function | | LatentOperationReLU | Apply ReLU function | | LatentOperationReLU6 | Apply ReLU6 function | | LatentOperationLeakyReLU | Apply LeakyReLU function | | LatentOperationELU | Apply ELU function | | LatentOperationSELU | Apply SELU function | | LatentOperationCELU | Apply CELU function | | LatentOperationGELU | Apply GELU function | | LatentOperationSiLU | Apply SiLU function | | LatentOperationHardSwish | Apply HardSwish function | | LatentOperationMish | Apply Mish function | | LatentOperationSoftplus | Apply Softplus function | | LatentOperationSoftmax | Apply Softmax function | | LatentOperationSoftmin | Apply Softmin function | | LatentOperationSoftsign | Apply Softsign function | | LatentOperationReshape | Change the shape of LATENT | | LatentOperationSlice | Extract a portion of LATENT | | LatentOperationRoll | Shift LATENT | | LatentOperationAddBroadcast | Add values with broadcasting | | LatentOperationMulBroadcast | Multiply values with broadcasting | | LatentOperationFill | Fill with specified value | | LatentOperationAdd | Add another LATENT | | LatentOperationMul | Multiply by another LATENT | | LatentOperationClamp | Limit values with upper and lower bounds | | LatentOperationClampMin | Limit values with lower bound | | LatentOperationClampMax | Limit values with upper bound | | LatentOperationApplyCFG | Apply CFG adjustment | | LatentOperationSplitCFG | Split CFG into conditional and unconditional parts | | LatentOperationInterpolate | Interpolate between two LATENT tensors | | Latent01ToImage | Convert LATENT in 0-1 range to image | | Latent11ToImage | Convert LATENT in -1 to 1 range to image (z = x * 0.5 + 0.5) | | GetSigma | Get σ at a specific index from SIGMAS |