@allyourbase
In spirit of the network school decided to share here a small research project
it's called helm
you feed it two prompts that contradict each other. it captures what changes between them, inside a single transformer block, and saves that as a tiny file — about 22 KB. i call it an "axis". then, when you sample, you dial it in with one number. 0 means the model is completely untouched, byte for byte.
it works on minimax h3 for video and z-image for images. no training run, no lora, the weights never move.
the reason it exists: prompts can't do "a little". change one word and the whole frame goes. i kept wanting a dial instead of a switch.
it's MIT licensed and it's on the comfy registry now.
github.com/U-N-B-R-A-N-D-E-D/comfyui-helm