sexual situations

physical violence

disturbing

male nudity

weapon violence

female swimwear or underwear

male swimwear or underwear

partial nudity

white supremacy

adult toys

graphic male nudity

illustrated explicit nudity

nudity

graphic violence or gore

wide hips

oral invitation

emaciated bodies

female nudity

undressed

male underwear

female swimwear

genitals

female underwear

thick thighs

breasts out

strapless leotard

vore

breast out

one breast out

huge breasts

gigantic breasts

huge butt

covered nipples

hair over breasts

no panties

sitting on face

anal

dildo riding

downblouse

oral

porn

futanari

hentai

nude

lingerie

sexy

nsfw

suggestive

self injury

extremist

hate speech

diapers

urine

incest

graphic female nudity

blowjob

child on child

hanging

hate symbols

nazi party

revealing clothes

pg-13

corpses

convenient censoring

peeing

exposed female nipple

sexual activity

sexual intent

latex clothing

swimwear

bukkake

fellatio

cumshot

implied fellatio

eat_cum

cumdrip

cum in pussy

cum on face

after fellatio

cum on hair

cum on body

cum on tongue

cum on hands

cum in mouth

triple fellatio

autofellatio

fucked silly

cum on pussy

pov fellatio

Retrained as a PEFT (i.e. standard) LoRA to increase compatibility with Nunchaku/SVDQuant variant of Flux-dev.

v8-flux

retrained using SimpleTuner in MMDiT-only mode:<pre><code>{
	"--resume_from_checkpoint": "latest",
 	"--data_backend_config": "/home/fox/LoRA/Blahaj/multidatabackend.json",
	"--aspect_bucket_rounding": 2,
	"--seed": 42,
	"--minimum_image_size": 0,
	"--disable_benchmark": false,
	"--output_dir": "output/blahaj-flux-v7-lycoris",
	"--lora_type": "lycoris",
	"--lycoris_config": "config/blahaj-flux/lycoris_config.json",
	"--lora_rank": 32,
	"--max_train_steps": 5000,
	"--flux_lora_target": "mmdit",
	"--num_train_epochs": 0,
	"--checkpointing_steps": 1000,
	"--checkpoints_total_limit": 5,
	"--model_type": "lora",
	"--pretrained_model_name_or_path": "black-forest-labs/FLUX.1-dev",
	"--model_family": "flux",
	"--train_batch_size": 1,
	"--gradient_checkpointing": "true",
	"--caption_dropout_probability": 0.0,
	"--resolution_type": "pixel_area",
	"--resolution": 512,
	"--validation_seed": 42,
	"--validation_steps": "500",
	"--validation_resolution": "1024x1024",
	"--validation_guidance": 5.0,
	"--validation_guidance_rescale": "0.0",
	"--validation_num_inference_steps": "20",
	"--validation_prompt": "blahaj toy plush shark",
	"--user_prompt_library": "/home/fox/LoRA/Blahaj/user_prompt_library.json",
	"--mixed_precision": "bf16",
	"--optimizer": "adamw_bf16",
	"--learning_rate": "1e-4",
	"--lr_scheduler": "polynomial",
	"--lr_warmup_steps": 100,
	"--text_encoder_1_precision": "no_change",
	"--text_encoder_2_precision": "no_change",
	"--max_grad_norm": 1.0,
	"--base_model_default_dtype": "bf16",
	"--quantize_via": "cpu",
	"--base_model_precision": "int8-quanto",
	"--validation_torch_compile": "false"
}
</code></pre>

v7-flux

Retrained based on <a target="_blank" rel="ugc" href="https://huggingface.co/OnomaAIResearch/Illustrious-xl-early-release-v0">https://huggingface.co/OnomaAIResearch/Illustrious-xl-early-release-v0</a>

v7-illustrious

Retrained using SimpleTuner in LyCORIS mode:<pre><code>{
	"--resume_from_checkpoint": "latest",
	"--data_backend_config": "/home/fox/LoRA/Blahaj/multidatabackend-1024.json",
	"--aspect_bucket_rounding": 2,
	"--seed": 42,
	"--minimum_image_size": 0,
	"--disable_benchmark": false,
	"--output_dir": "output/blahaj-sdxl-v7",
	"--lora_type": "lycoris",
	"--lycoris_config": "config/blahaj-flux/lycoris_config.json",
	"--max_train_steps": 5000,
	"--num_train_epochs": 0,
	"--checkpointing_steps": 500,
	"--checkpoints_total_limit": 5,
	"--model_type": "lora",
	"--lora_rank": 32,
	"--pretrained_model_name_or_path": "stabilityai/stable-diffusion-xl-base-1.0",
	"--model_family": "sdxl",
	"--train_batch_size": 1,
	"--gradient_checkpointing": "true",
	"--caption_dropout_probability": 0.0,
	"--resolution_type": "pixel_area",
	"--resolution": 1024,
	"--validation_seed": 42,
	"--validation_steps": 500,
	"--validation_resolution": "1024x1024",
	"--validation_guidance": 4.2,
	"--validation_guidance_rescale": "0.0",
	"--validation_num_inference_steps": "20",
	"--validation_prompt": "blahaj toy plush shark",
	"--user_prompt_library": "/home/fox/LoRA/Blahaj/user_prompt_library.json",
	"--mixed_precision": "bf16",
	"--optimizer": "adamw_bf16",
	"--learning_rate": "1e-4",
	"--lr_scheduler": "polynomial",
	"--lr_warmup_steps": 100,
	"--base_model_precision": "int8-quanto",
	"--validation_torch_compile": "false"
}</code></pre>

v7-sdxl

LyCORIS version trained against PonyXL

v7-pony

Trained against SD3.5-Large<pre><code>{
	"--resume_from_checkpoint": "latest",
	"--data_backend_config": "/home/fox/LoRA/Blahaj/config/blahaj-sd35/multidatabackend.json",
	"--aspect_bucket_rounding": 2,
	"--seed": 42,
	"--minimum_image_size": 0,
	"--disable_benchmark": false,
	"--output_dir": "output/blahaj-sd35-v7",
	"--lora_type": "lycoris",
	"--lycoris_config": "config/blahaj-sd35/lycoris_config.json",
	"--lora_rank": 32,
	"--max_train_steps": 5000,
	"--num_train_epochs": 0,
	"--checkpointing_steps": 1000,
	"--checkpoints_total_limit": 5,
	"--model_type": "lora",
	"--pretrained_model_name_or_path": "stabilityai/stable-diffusion-3.5-large",
	"--model_family": "sd3",
	"--train_batch_size": 1,
	"--gradient_checkpointing": "true",
	"--caption_dropout_probability": 0.0,
	"--resolution_type": "pixel_area",
	"--resolution": 1024,
	"--validation_seed": 42,
	"--validation_steps": "500",
	"--validation_resolution": "1024x1024",
	"--validation_guidance": 5.0,
	"--validation_guidance_rescale": "0.0",
	"--validation_num_inference_steps": "20",
	"--validation_prompt": "blahaj toy plush shark",
	"--user_prompt_library": "/home/fox/LoRA/Blahaj/user_prompt_library.json",
	"--mixed_precision": "no",
	"--optimizer": "adamw_bf16",
	"--learning_rate": "1e-4",
	"--lr_scheduler": "polynomial",
	"--lr_warmup_steps": 100,
	"--base_model_default_dtype": "bf16",
	"--quantize_via": "cpu",
	"--max_grad_norm": 0.01,
	"--flux_schedule_shift": 1,
	"--base_model_precision": "int8-quanto",
	"--text_encoder_1_precision": "no_change",
	"--text_encoder_2_precision": "no_change",
	"--text_encoder_3_precision": "no_change",
	"--validation_torch_compile": "false"
}
</code></pre>

v7-sd35

This seems to be the worst version. Maybe I did something wrong when training or simply because SD3.5 is, well, like that. Reducing LoRA weight to 0.8 might help a bit.

v7-sd35-medium

retrained with network_dim 32 (was 4)

v6-flux

v5-flux

a few more images, and less steps which would hopefully improve flexibilitystill no idea what i'm doing

Type	LoRA
Stats	31 0
Reviews	Positive (7)
Published	Apr 20, 2025
Base Model	Flux.1 D
Training	Steps: 2,500
Trigger Words	blahaj shark
Hash	AutoV2 6245D4084B

Blahaj!