Skip to main content
The LotusConditioning node provides pre-computed conditioning embeddings for the Lotus model. It uses a frozen encoder with null conditioning and returns hardcoded prompt embeddings to achieve parity with the reference implementation without requiring inference or loading large tensor files. This node outputs a fixed conditioning tensor that can be used directly in the generation pipeline.

​
Inputs

ParameterDescriptionData TypeRequiredRange
No inputsThis node does not accept any input parameters.---

​
Outputs

Output NameDescriptionData Type
conditioningThe pre-computed conditioning embeddings for the Lotus model, containing fixed prompt embeddings and an empty dictionary.CONDITIONING
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub
Source fingerprint (SHA-256): 950e09b7fda911c7596445cf1a90ec10f451b396534f149b74e86b2da7deacc0