Skip to main content
Turn the interleave history left by SenseNova Conditioning into usable text, images and thinking. Run it at the end of the loop, with End Loop accumulate disabled, to collect everything the model produced in one pass.

Inputs

Constraints and behavior:
  • The conditioning must come from SenseNova Conditioning in the interleave or self-correction task; anything else raises an error.
  • When the model is waiting for an image, connect the sampler’s LATENT output to samples.
  • Reference images connected at the start of the run are excluded from the output, and image consumers are skipped when no image was generated.

Outputs

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): 62c7ded50c1348e2242319209729a17d543d80e4d01ddf173cf8b77e4fd0be50