









Flux training can be quite expensive so if you enjoy this LORA (which I decided to NOT release as Early Access) please consider donating any amount of buzz, give a like, post a review directly to the model page, share with a friend, leave a comment, buy my badge from the shop or do whatever you want to support me. Thanks
No trigger needed but I've found it helps guide FLUX and the LORA if you make your thing "metallic" or "chrome" or whatever else you can thing of that tells FLUX what you want changed based on the LORA's training, have a look at the examples that show the LORA on/off.
Weight: ? - 1 (recommended 1)
Well, here's a first, posting 3 versions of this LORA and I'm potentially working on a 4th. As an experiment they were trained on mostly the same dataset but with different or no tagging.
I highly recommend v3 based on my tests but if you're unsure which one to download you can always test the same prompt with all of them in the Civit online generator.
Triggers: dvr-chrm
Weight: 0.25 - 1 (depending on the effect you're after, start with 0.8)
If you haven't already, come get your Ds and check out my SDXL model merge (Dever's Divine Diffusion or DDD).
If you want to support me and my work make sure to submit any reviews and images directly to the LORA or MODEL page. Cross-posts do not currently award the creator any sort of buzz, this includes posting from the Create menu at the top (not cool, Civit please fix).
Alternatively you can purchase my badge called "Attention to details" in the CivitAi shop. Big badge warning :)

I tried my best to highlight some of the possible use cases and interesting ways to use the LORA.
I hope you like it and remember to post your creations directly to the LORA page if possible.
Thank you <3
1. 轉载模型僅供學習與交流分享,其版權及最終解释權归原作者。
2. 模型原作者如需認領模型,請通過官方渠道联系海藝AI工作人員進行認證。我們致力於保護每一位創作者的權益。 點擊去認領
