Verified: 7 months ago
SafeTensor
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.
First try to contain the style of the Mononoke anime series in a LoRa, also first real attempt to create a style LoRa in Flux. This attempt was made without using captions, though I might make another attempt in the near future, I think captioning people and a slightly higher number of training steps would have given me better results. I did like the endresult somewhat so I am still going to upload it! (I didnt want to waste 3 hours of training xD)
As a trigger I would recommend using, "in the style of m0n0n0k3" and adding "anime" to it might help too. You can go pretty wild with the strength, I used 1 for my images, but going up to 1.4 - 1.5 still creates decent results imo, you get more of a "smudged" look. The image with the bicycle for example is 1.3 strength. Lower strengths than 1 still give a cellshaded look but the characters tend to go a bit more towards the realistic.
Edit: i did notice a tendency to generate female faces, I think this has to do with zero captioning. Will try to fix this in a later version.