models/MJanime_Flux_LoRa_v3_Final - MJanime_Flux_LoRa_v3.1

MJanime_Flux_LoRa_v3_Final - MJanime_Flux_LoRa_v3.1

9/11/2025

11:23:26 PM

Related Keywords & Tags

anime,anime style,flux.1 d,flux anime style,flux lora model,lora,mjanime_flux_lora_v3.1,mjanime_flux_lora_v3_final,stable diffusion,stan_katayama,style

A fearsome rabbit dressed in rugged armor stands in a wasteland, holding a glowing sword with determination, against a dramatic sky with dynamic shadows.

A scarred white rabbit wearing damaged, bloody knight armor lies sleeping peacefully on a brightly lit meadow with red flowers, set against a dark and desolate wasteland backdrop.

Close-up of a punk-styled woman with messy black hair, wearing a black choker and strapless top, illuminated by vibrant neon orange and green lights in a cyberpunk cityscape background.

Digital painting of a red ghostly humanoid figure with a skeletal translucent appearance, shining against a dark background.

Dark moody painting of a mysterious tree-like structure with twisted leafless branches and thatched roof atop a moss-covered rock in a foggy stormy landscape, illuminated by dim eerie light with heavy clouds and flying birds in the background.

Detailed realistic depiction of a skeleton lying on a grassy hill with mountainous ruins and a cloudy overcast sky in the background.

Aerial view of a person standing on a crater edge, looking into a dark abyss where an eerie monstrous face emerges.

Tips

Load LoRa models in fp16 if possible for better performance

For a pronounced anime style, use strength values between 0.90 and 1.30

Try varying the strength to around 0.5 or 0.6 for more realistic image outputs

The model was trained with the optimizer "prodigy" in rank 32

The model was created with Ostris script available at https://github.com/ostris/ai-toolkit?tab=readme-ov-file#flux1-training

Creator Sponsors

All models are exclusive to Civitai! Unauthorized publication will be reported.

Hello everyone!

All models are exclusive to Civitai ! Anyone who publishes my models without my consent will be reported!

Trigger Word : Anime style

I highly recommend loading my lora models in fp16 if possible

Since Flux's anime style is a bit weak, I decided to tackle it for my first training of Flux LoRa model.

v3_Final : This version uses the same dataset as v3.1 but this time at 100 epoch against 16 before.

The poses are more dynamic and the characters more expressive

V3.1 : A new dataset of 330 images. Better selection so that characters look older if requested.
The overall quality is also much better

V2_Final : same as v2_15k, but the style is much more pronounced

v2_15k : This version has a dataset of 338 images against 112 for version 1.

This is not the final version of this v2, but I find this 15k steps version very interesting and rather stable.

This version was trained with the optimizer "prodigy" in rank 32

The style is there, but the final version will be very different!

V1 : I used for this 112 high quality images generated with MidJourney v6 in the --personalize style.

Be indulgent, this is a first try. If you zoom in a bit on the images the lines can appear a bit pixelated. But the style is pretty cool!

For a pronounced anime style, Sweet spot between 0.90 and 1.30

I recommend you to play with the strength to vary the pleasures for example at 0.5 or 0.6 by asking for a realistic image.

The model was created with Ostris script which you can find here :
https://github.com/ostris/ai-toolkit?tab=readme-ov-file#flux1-training

Have fun! 😊