After creating the Schnell LoRA for Flux.1 D - v1.0 | Stable Diffusion LoRA | Civitai I asked myself... what if I extracted the Dev model from the Schnell model? Could this result in a Dev LoRA that retains some unique properties from the Schnell model when applied to a Schnell model?

Turns out, it does! When you apply this LoRA at the end of your LoRA chain on a Schnell model, it enables you to produce 20 steps outputs from a Flux.1 S model... and you can combine other LoRAs if you like... but with not so perfect results... need more experimentation with that.

With this LoRA, you can't specify the Flux Guidance, and the results are not as nice... but since I made it... I figure, let's share it.

Experiment and enjoy!