Comfy.ICU
ExploreDocsPricing
Support

ComfyUI Node: Init SD3 LoRA Training

Authored by kijai

Created

Updated

781 stars

Category

FluxTrainer/SD3

Inputs

sd3_models TRAIN_SD3_MODELS
dataset JSON
optimizer_settings ARGS
output_name STRING
output_dir STRING
network_dim INT
network_alpha FLOAT
learning_rate FLOAT
max_train_steps INT
cache_latents
  • disk
  • memory
  • disabled
cache_text_encoder_outputs
  • disk
  • memory
  • disabled
training_shift FLOAT
highvram BOOLEAN
blocks_to_swap INT
fp8_base BOOLEAN
gradient_dtype
  • fp32
  • fp16
  • bf16
save_dtype
  • fp32
  • fp16
  • bf16
  • fp8_e4m3fn
  • fp8_e5m2
attention_mode
  • sdpa
  • xformers
  • disabled
train_text_encoder
  • disabled
  • clip_l
  • clip_l_fp8
  • clip_l+T5
  • clip_l+T5_fp8
clip_l_lr FLOAT
clip_g_lr FLOAT
T5_lr FLOAT
sample_prompts STRING
gradient_checkpointing
  • enabled
  • disabled
additional_args STRING
resume_args ARGS
block_args ARGS
loss_args ARGS

Outputs

NETWORKTRAINER

INT

KOHYA_ARGS

Extension: ComfyUI Flux Trainer

Currently supports LoRA training, and untested full finetune with code from kohya's scripts: a/https://github.com/kohya-ss/sd-scripts

Authored by kijai

View Nodes

Run ComfyUI workflows in the Cloud!

No downloads or installs are required. Pay only for active GPU usage, not idle time. No complex setups and dependency issues

Learn more