Generated image not correct
Hello,
I have moved the shenhe lora in correct folder and even imported it in additional networks in txt2img as shown in screenshot but I am not able to get results no where close to the example from which I copied the prompt.
Any help would be much appriciated
Is that Stable Diffusion 1.5? Then it definitely won't work as original stable diffusion does not train with Danbooru tags, so you need to modify the prompt a bit so that it is not based on Danbooru tags or import another custom models.
Oh I see, do I have to use some other version of stable diffusion so it accepts Danbooru tags? If so, can you link me the repo from which I can download it.
Realistic models: https://huggingface.co/nuigurumi/basil_mix, https://civitai.com/models/6925/realdosmix (There are other models too, but I don't know a lot of realistic models...)
Anime models: https://huggingface.co/WarriorMama777/OrangeMixs, https://huggingface.co/Linaqruf/anything-v3.0, https://huggingface.co/andite/anything-v4.0, and many others... most of the anime models support Danbooru.
I finally managed to figure out the models and how they work.
How can I know what model you used in your sample prompts?
Sorry but I'm not the developer so I think you better ask
@YoungMasterFromSect
for that...
I personally think he used AbyssOrangeMix series
Okay thanks a lot!!