How to train a lora for pipeline_qwenimage_edit_plus?

[Xiangyu33]

I'm new to diffusers, I have noticed that qwenimage_edit is released in diffusers pipeline

Could anyone plz tell me how to train a lora for it ? looking for kind reply :)

Is it right to run scripts below?

export MODEL_NAME="Qwen/Qwen-Image-Edit-2509"
export DATASET_NAME="***" // my dataset name

accelerate launch --mixed_precision="fp16" train_text_to_image_lora.py \
  --pretrained_model_name_or_path=$MODEL_NAME \
  --dataset_name=$DATASET_NAME --caption_column="text" \
  --resolution=512 --random_flip \
  --train_batch_size=1 \
  --num_train_epochs=100 --checkpointing_steps=5000 \
  --learning_rate=1e-04 --lr_scheduler="constant" --lr_warmup_steps=0 \
  --seed=42 \
  --output_dir="sd-naruto-model-lora" \
  --validation_prompt="cute dragon creature" --report_to="wandb"

[Xiangyu33]

should i use train_text_to_image_lora.py to train lora of Qwen-image-edit-2509? I'm not sure

[asomoza]

we don't have a training script for Qwen-image-edit-2509 at the moment. The only option with diffusers you have right now, is to maybe adapt the kontext training script to qwen image edit.

[Xiangyu33]

thanks for kind reply, however, i notice that in #11831, diffusers doesn't supply a scripts for kontext lora training. should i look lora training scripts in other libs?

[asomoza]

Hi, the i2i training for kontext was added at a later date than the answer, we now have it here