Generate cool images with an adventurous feel using this embedding!
This embedding can create both beautiful landscapes and cool characters with a fantasy/adventurous feeling.
Tested with Dreamshaper & Dreamlike Diffusion
Include "landscape" in your prompt to emphasize you want a landscape, and "character" to emphasize you want a character.
Please share what you make, I love seeing it all!
Description
Trained to 2300 steps on a dataset of 25 images.
FAQ
Comments (13)
share them, son
can you share your training configuration please?
I think is was around 25-30 hand captioned images, 512x512, batch size of 5, gradient accumulation of 1, 2,300 steps (it probably went to around 10000 but I saved every 50 or 100 and selected the best of those), and 460 epochs.
@junglerally_ thank you for sharing!!
hi can i sell the image generated from this emmbedings?
I think most of these 'Licenses', when the creator sets them out/informs about them, are for the model itself, not for the images generated...
@thiagojramos ohhh cause i dont see there have any label on this model , some will label cant sell the image that geneated for this model something like that
@yinquest11 And how the hell is someone supposed to know the model(s) used in the image? Besides those parameters in the image (which can just be overwritten, even just by saving it again in any image viewer/editor they are removed) I don’t know how. I understand having a license for the use of the model, but for the generated image? 🤷♂️
@thiagojramos haha some model have very strong style such as 墨心 MoXin - 墨心 MoXin 1.0 | Stable Diffusion LoRA | Civitai
it can be see through , so if the creator not allow then i just wont sell la , that is more save
Oh, this comment reeks of unoriginality
I love this Lora. Making beautiful landscapes is simply... addicting!
Is it possible to get a version for SDXL? Thank you
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.