I am looking for workflows that are the most basic core structures for generating i1i it2i locally with models I can run on comfyui. Because there are so many different models, or architectures of models along with the loras to complement them, its been a bit challenging to learn how to use this new technology. There are so many sources on youtube, i dont know what the good educating ones are . . .
I have been experimenting with guidance assistance from a local AI llm, which has been very helpful, its just a little old and not the best educator.
I was hoping to find a basic workflow for each i2i and it2i for a few different models, anima, sd1.5, sdxl, pony, illustrious, flux, kr2 and qwen . . . These I wanna start with and learn to adapt workflows around the models and the loras that those specific models use.
Can anyone point me in the direction to this type of specific information? I am trying to better understand the structures for the different types in their most basic form,
I have I think, sd1.5 so far figured out for the most basic part. I am able to generate images of what I want in the prompt, I can add a lora, use that effectively, I also setup a second image input to influence the result mixed from the prompt. But learning and progress has been slow, I dont have much time to spend searching YT or asking misguided questions in AI.
Any help would be appreciated.
Ideally, I wanna see the basic core nodes and how they connect for each different type of model type and its relative parts. I want to use only two loras each workflow (model based) because I plan on training loras, and I plan on using controlNET once I learn that . . . I think that will be what I want to control subjects in my art.
I have been experimenting with guidance assistance from a local AI llm, which has been very helpful, its just a little old and not the best educator.
I was hoping to find a basic workflow for each i2i and it2i for a few different models, anima, sd1.5, sdxl, pony, illustrious, flux, kr2 and qwen . . . These I wanna start with and learn to adapt workflows around the models and the loras that those specific models use.
Can anyone point me in the direction to this type of specific information? I am trying to better understand the structures for the different types in their most basic form,
I have I think, sd1.5 so far figured out for the most basic part. I am able to generate images of what I want in the prompt, I can add a lora, use that effectively, I also setup a second image input to influence the result mixed from the prompt. But learning and progress has been slow, I dont have much time to spend searching YT or asking misguided questions in AI.
Any help would be appreciated.
Ideally, I wanna see the basic core nodes and how they connect for each different type of model type and its relative parts. I want to use only two loras each workflow (model based) because I plan on training loras, and I plan on using controlNET once I learn that . . . I think that will be what I want to control subjects in my art.