ComfyUI Node: Train LoRA

Authored by comfyanonymous

Created

Updated

86154 stars

Category

training

Inputs

model MODEL
latents LATENT
positive CONDITIONING
batch_size INT
grad_accumulation_steps INT
steps INT
learning_rate FLOAT
rank INT
optimizer
  • AdamW
  • Adam
  • SGD
  • RMSprop
loss_function
  • MSE
  • L1
  • Huber
  • SmoothL1
seed INT
training_dtype
  • bf16
  • fp32
lora_dtype
  • bf16
  • fp32
algorithm
  • LoRA
  • LoHa
  • LoKr
  • OFT
gradient_checkpointing BOOLEAN
existing_lora
  • [None]

Outputs

MODEL

LORA_MODEL

LOSS_MAP

INT

Extension: ComfyUI

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.

Authored by comfyanonymous

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