Description
Variant 2 is a merge of Variant 1 with Anime art and Mangaka 3012. This gave more details and a widening of the graphic field.
Weighting 1.5 but you can test other setting.
Good tags : in style of comic book, an illustration of.
FAQ
Comments (22)
That's cool! That's great! (❁´◡`❁) (●'◡'●)
One sentence "Ohh my GOD" - amazing job. Thank you.
Thank you very much for the wonderful model. Good luck in your future creative work!
Thanks
great lora I'm loving it but there is one thing I'm running into I was hoping you could tell me how I might fix it. I keep trying to generate african americans but it does not seem to work. It just defaults to white people or at least non dark skinned people
the training image base did not include ethnicity variants. Very few retro images took this variety into account. This might be fixed in a later release.
@Adel_AI I'd be very grateful if you did. Thank you for answering my question.
Maybe instead of "African American", you can try "Black African". The word "American" may bias the model toward white Americans with this LoRA. Here is one working example: https://civitai.com/images/35354379
If you still have problem producing Black Africans, post your prompt, and we can take a look at it.
Thanks for the Lora. How do you guys merge loras?
Thanks. Merge with https://github.com/larsupb/LoRA-Merger-ComfyUI
cool
Wow! Nice👍
Variant 2 is insane, I'm screaming
Amazing Lora! Thank you! A shame it doesn't work in Invoke. Would be great if that could be fixed...
Thank you for your feedback. Unfortunately, I haven't used Invoke for a long time, so I won't be able to find out the origin of this problem and therefore resolve it. I read comments on the web reporting the malfunction of invoke with loras, it seems that the problem is generalized and not linked to my model
@Adel_AI My experience with Invoke and loras is great sofar. Only a couple are not working. All of them Flux loras though. Maybe it will change with a future release. Also dropped the problem with them. But it's no big issue. There are other ways to use it. Thanks again for your work! Great fun!
Hello I wanted to say there is a guy that is using your lora for credits on seaart. Pls do something.
thank you for your feedback, unfortunately there is absolutely nothing we can do about this abuse, the sites which host them show total complicity given that this makes their platform run. I have reported this abuse several times without ever obtaining its deletion.
@Adel_AI What if you claim it ? Claim the model .
Extremely flexible, fairly accurate and easy to use. Caters to a wide variety of settings and compositions.
Versatile enough
Hello! Please tell me if this LoRA will work with the flux1-dev-bnb-nf4-v2 model?
Details
Files
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.