Network Bending for ComfyUI
A custom node pack for ComfyUI that enables creative manipulation and 'bending' of neural network models. Perform various operations on loaded model checkpoints to create unique and experimental effects.
Nodes (18)
The node that turns a waveform into something you can wire into a visual effect
Audio Latent Blend is deprecated — here's what it did and what replaced it
Steer an audio diffusion model toward a reference sound
Morph one sound into another in the latent space
An effects rack for audio latents, not audio files
Turn up to four reference tracks into one steerable latent
Put the texture of one sound onto another
Bring an audio latent back to a waveform you can hear
The front door to the audio latent space
Prune the wrong weights on purpose
The bridge between image latents and audio latents
Merge two checkpoints without leaving your workflow
The 'break it on purpose' node that started this pack
The advanced node that's honest about being unfinished
Rearrange the channels of a latent like they're audio faders
Corrupt the latent itself, leave the model alone
Blend two VAEs like you'd blend two checkpoints
Surgery on the VAE itself, not the images it produces
Network Bending for ComfyUI
A custom node pack for ComfyUI that enables the creative manipulation of generative models. Allows you to corrupt, modify, and blend neural networks.
Features
- Add Noise: Inject Gaussian noise into model weights
- Scale Weights: Scale model weights up or down to amplify or dampen features
- Prune Weights: Remove small weights for sparsification
- Randomize Weights: Replace portions of weights with random values
- Smooth Weights: Apply spatial smoothing to weight matrices
- Quantize Weights: Reduce weight precision to discrete levels
- Layer Swap: Swap weights between similar layers
- Activation Replace: Replace activation functions
- Weight Transpose: Transpose weight matrices
- Channel Shuffle: Randomly shuffle channels in convolutional layers
- Frequency Filter: Apply frequency domain filtering
- Weight Clustering: Group similar weights together
Mix two models together with various blending modes:
- Linear Interpolation: Simple weighted average of weights
- Weighted Sum: Weighted sum with normalization
- Layer-wise Mix: Different mix ratios for different layers
- Frequency Blend: Mix models in frequency domain
- Random Mix: Randomly select weights from either model
Audio Processing Nodes
The package also includes comprehensive audio processing capabilities for Stable Audio models:
Audio VAE Nodes
- Audio VAE Encode: Encode audio waveforms into latent space
- Audio VAE Decode: Decode latent representations back to audio
Audio Latent Manipulation
- Audio Latent Interpolate: Smoothly interpolate between audio latents
- Audio Latent Blend: Blend multiple audio latents with various modes
- Audio Latent Manipulator: Direct manipulation of audio latent features
- Audio Feature Extractor: Extract specific features from audio latents
Audio Style Transfer
- Audio Style Transfer: Transfer style from one audio to another
- Audio Latent Guidance: Guide generation with reference audio
- Audio Reference Encoder: Encode reference audio for style transfer
Installation
Basic Installation
- Clone this repository into your ComfyUI
custom_nodesdirectory:
cd ComfyUI/custom_nodes
git clone https://github.com/DavidPiazza/network_bending.git
- Restart ComfyUI
Audio Network Bending
- Load an audio model (e.g., Stable Audio VAE)
- Add audio processing nodes from the network_bending/audio category
- Connect audio inputs either from file loaders or generated audio
- Apply bending operations to the audio latents
- Decode back to audio using Audio VAE Decode
Cross-Modal Bending
Apply image model bending techniques to audio models or vice versa for experimental results.
License
This project is licensed under the GNU General Public License v3 (GPL-3.0) - see the LICENSE file for details.