- More cool models.
ℹ️ This model comes with new finetuning and a new recipe. Includes ❤️🔥 Divas and 💖 Babes 2.0.
ℹ️ Trigger words from ❤️🔥 Divas and 💖 Babes 2.0 can be used.
Description
ℹ️ Added a little more sassiness with a few additional finetunings.
ℹ️ Trigger words from ❤️🔥 Divas and 💖 Babes 2.0 might work.
FAQ
Comments (5)
I'm giving Civitai generation service another chance. Hopefully, users will try to understand how to use SD before reviewing models.
[Original message:]
Civitai generation service has been blocked due to inexperienced users using it to generate low-quality images and leaving bad reviews.
???
But there's only one negative (and very sketchy-looking) review on this model... :/
@Anonimous1234567890
Yes, the problem is more general, I blocked the Civitai generation service for all my checkpoints.
I'm not feeling very good about it. However, I had a few interactions with users who started using the service the same day, and they think that the small and low-quality images that the service generates is all that the model has to offer.
Users who generate images locally learn how to use hires-fix and improve the quality of images with prompting. Unforantly quite a few people who use Civitia generation service lately, jump to judging before they know how to use SD.
I don't feel it's fair to my models or to me. I invested a lot of time and effort in developing my models, and I released them for free, so the minimum that I want for them is to receive fair reviews.
Other model creators have similar experiences and some of them blocked generation service too.
[@alexds9]
It's interesting to know. I didn't know about the poor quality. I'm glad to know that I was right to use SD locally.
I like to be thorough with what I do, I'm still new to SD.
But I can say with certainty that creating these models is an odyssey.
I wish I could use hire-fix (insufficient varm :/)
@juanluix
Thank you very much for your support. ❤️️
I think that Civitai trying to improve the service and hopefully they will introduce hires-fix eventually.
There is on demand service that I use for training - vast.ai it has a large selection of servers with GPUs, including servers that would work really well for prompting. I checked the price per hour for a server with RTX 3060 12GB Vram, and it's around 0.065 USD. So for 1 USD you can use the server for around 15 hours. If you use it for SD for 4 hours every day of the month, it will cost you around 8 USD.



















