Monody FLUX

Monody FLUX

The model aims to create cute and adorable characters with soft pastel colors and aesthetic anime style

Currently still in Beta stage, this model still cannot perform very complex prompts

________________________________________________________________________________________________________

UPDATE :

The Beta version has major improvements

- add Aesthetic shadow

- sharpens the colors more vivid

- fixing background & clothing

- work with natural language

- improve dynamic angle & pose

- improve skin texture & nsfw

________________________________________________________________________________________________________

Usage Recommendation

Sampler : Euler

Steps : 20+

CFG scale

Comfyui : 1 - 2

Shakker Generator : 3 - 7

A1111 Shakker : 1 - 5

Resolution

768x1152 (Portrait)

1152x768 (Landscape)

Quality Prompt : masterpiece, aesthetic

Negative Prompt :

________________________________________________________________________________________________________

Comparison

The test was done using Comfyui with the same prompts, seed, and settings

_{"An animated image of an anime girl with long white hair and blue eyes. She is sitting on her knees on a grassy field. She has a blue and black bow in her hair. There are blue and white butterflies in the grass surrounding her. There is a large white moon in the sky above her head. To the right of her is a tall tower with a pointed top."}

_{"masterpiece, 1girl, holding ice cream cone, confetti, multiple balloons, indoors, joyful, happy, dress"}

_{"An animated image of a girl with long blonde hair and brown eyes. She is wearing a brown coat with white fur trim and brown gloves. The girl is standing on a snow covered ground. To the left of the girl is a brown building with a black lamp on the side of it. The sky is blue with white fluffy clouds."}

_{"masterpiece, 1girl, riding bycycle, school uniform, black serafuku, looking at viewer, smile, medium hair, outdoors, blue sky, meadow"}

_{"An animated image of a girl with red hair sitting on a black container. The girl is wearing a white short-sleeved shirt and white tennis shoes. The cat is sitting on the left side of the image. The black cat has a black tail with white whiskers. The sky is dark blue with a white moon in the upper left corner. The moon is partially visible in the top right corner. There is a window to the right of the girl."}

_{"masterpiece, 1girl, blush, open mouth, brown boots, leaning forward, looking at viewer, pov, forest, flowers, blue sky, dynamic angle, hand on knee, leather backpack, safari hat, explorer outfit"}

________________________________________________________________________________________________________

Conclusion

This model is still very limited in what it can do, with not many aspect ratios in the dataset, it will make it difficult for this model to handle ratios outside the limits.

Sometimes bad anatomy and inaccurate often appear, this is due to overfitting in the model.

Some complex prompts are not recognized by this model, you have to write them specifically.

This model has not been tested with LoRa trained from the FLUX base model.

Please review with images, it will be very useful for improving the next version, thank you

________________________________________________________________________________________________________