Skip to main content
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub
The ModelMergeLTXV node performs advanced model merging operations specifically designed for LTXV model architectures. It allows you to blend two different models together by adjusting interpolation weights for various model components including transformer blocks, projection layers, and other specialized modules.

Inputs

ParameterData TypeRequiredRangeDescription
model1MODELYes-The first model to merge
model2MODELYes-The second model to merge
patchify_proj.FLOATYes0.0 - 1.0Interpolation weight for patchify projection layers (default: 1.0)
adaln_single.FLOATYes0.0 - 1.0Interpolation weight for adaptive layer normalization single layers (default: 1.0)
caption_projection.FLOATYes0.0 - 1.0Interpolation weight for caption projection layers (default: 1.0)
transformer_blocks.0.FLOATYes0.0 - 1.0Interpolation weight for transformer block 0 (default: 1.0)
transformer_blocks.1.FLOATYes0.0 - 1.0Interpolation weight for transformer block 1 (default: 1.0)
transformer_blocks.2.FLOATYes0.0 - 1.0Interpolation weight for transformer block 2 (default: 1.0)
transformer_blocks.3.FLOATYes0.0 - 1.0Interpolation weight for transformer block 3 (default: 1.0)
transformer_blocks.4.FLOATYes0.0 - 1.0Interpolation weight for transformer block 4 (default: 1.0)
transformer_blocks.5.FLOATYes0.0 - 1.0Interpolation weight for transformer block 5 (default: 1.0)
transformer_blocks.6.FLOATYes0.0 - 1.0Interpolation weight for transformer block 6 (default: 1.0)
transformer_blocks.7.FLOATYes0.0 - 1.0Interpolation weight for transformer block 7 (default: 1.0)
transformer_blocks.8.FLOATYes0.0 - 1.0Interpolation weight for transformer block 8 (default: 1.0)
transformer_blocks.9.FLOATYes0.0 - 1.0Interpolation weight for transformer block 9 (default: 1.0)
transformer_blocks.10.FLOATYes0.0 - 1.0Interpolation weight for transformer block 10 (default: 1.0)
transformer_blocks.11.FLOATYes0.0 - 1.0Interpolation weight for transformer block 11 (default: 1.0)
transformer_blocks.12.FLOATYes0.0 - 1.0Interpolation weight for transformer block 12 (default: 1.0)
transformer_blocks.13.FLOATYes0.0 - 1.0Interpolation weight for transformer block 13 (default: 1.0)
transformer_blocks.14.FLOATYes0.0 - 1.0Interpolation weight for transformer block 14 (default: 1.0)
transformer_blocks.15.FLOATYes0.0 - 1.0Interpolation weight for transformer block 15 (default: 1.0)
transformer_blocks.16.FLOATYes0.0 - 1.0Interpolation weight for transformer block 16 (default: 1.0)
transformer_blocks.17.FLOATYes0.0 - 1.0Interpolation weight for transformer block 17 (default: 1.0)
transformer_blocks.18.FLOATYes0.0 - 1.0Interpolation weight for transformer block 18 (default: 1.0)
transformer_blocks.19.FLOATYes0.0 - 1.0Interpolation weight for transformer block 19 (default: 1.0)
transformer_blocks.20.FLOATYes0.0 - 1.0Interpolation weight for transformer block 20 (default: 1.0)
transformer_blocks.21.FLOATYes0.0 - 1.0Interpolation weight for transformer block 21 (default: 1.0)
transformer_blocks.22.FLOATYes0.0 - 1.0Interpolation weight for transformer block 22 (default: 1.0)
transformer_blocks.23.FLOATYes0.0 - 1.0Interpolation weight for transformer block 23 (default: 1.0)
transformer_blocks.24.FLOATYes0.0 - 1.0Interpolation weight for transformer block 24 (default: 1.0)
transformer_blocks.25.FLOATYes0.0 - 1.0Interpolation weight for transformer block 25 (default: 1.0)
transformer_blocks.26.FLOATYes0.0 - 1.0Interpolation weight for transformer block 26 (default: 1.0)
transformer_blocks.27.FLOATYes0.0 - 1.0Interpolation weight for transformer block 27 (default: 1.0)
scale_shift_tableFLOATYes0.0 - 1.0Interpolation weight for scale shift table (default: 1.0)
proj_out.FLOATYes0.0 - 1.0Interpolation weight for projection output layers (default: 1.0)

Outputs

Output NameData TypeDescription
modelMODELThe merged model combining features from both input models according to the specified interpolation weights