Verified: 2 years ago
SafeTensor
Created on Civitai
The FLUX.1 [dev] Model is licensed by Black Forest Labs. Inc. under the FLUX.1 [dev] Non-Commercial License. Copyright Black Forest Labs. Inc.
IN NO EVENT SHALL BLACK FOREST LABS, INC. BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.
Trigger Word: l0lspx style
League of Legends Splash Art Style LoRA
This model is trained on CivitAI using 350 League of Legends splash art images, with a total of 5250 steps.
It helps to create the artstyle of League of Legends champion splash arts.
No champion names were included during training, so there is no reason to include them in your prompts!
Note: No LoRA's other than this were used during the generation of sample images. Combining different LoRA's may result in higher quality images.
Important Tips:
The recommended weight is: 0.6-0.8 in most cases. However, on very rare occasions higher values up to 1.6 may also work.
Try placing the trigger word "l0lspx style" at the beginning or within the first sentence of your prompt. You may also experiment with including the trigger word in each sentence.
I recommend a FluxGuidance value of 1.5-2.5. But feel free to experiment with higher values to see what works best for your needs.
If you are going to upscale with a second pass, try reducing the LoRA weight and increasing the FluxGuidance just a bit.
Since the entire dataset consists of 16:9 images, it can create very good wallpaper images with the same aspect ratio. It works well on all base Flux resolutions.
And the second one is WITH my LoRA