LoRA Training

Train an SDXL LoRA

Fine-tune a LoRA adapter on Stable Diffusion XL — the most common LoRA training target for style and subject customization in ComfyUI workflows.

Estimated $2–$8 per training run · 15–45 min

What inputs you need

  • 10–30 training images
  • Captions (auto or manual)
  • Training config (rank, steps, learning rate)

What Badgr returns

  • Trained LoRA safetensors file
  • Training loss curve / checkpoint log
  • Sample generations at checkpoints

Recommended GPU routes

Estimated cost: $2–$8 per training run · Estimated runtime: 15–45 min

Example command

badgr run "python train_lora.py --base sdxl --config sdxl_lora.yaml" \
  --gpu A100_40GB \
  --max-cost 8 \
  --max-runtime 60

Common setup failures Badgr avoids

Overfitting on a small dataset — default config caps steps relative to image count and logs sample generations each checkpoint so you can catch it early.

Dataset upload timeout on job start — pre-stage images to object storage and pass a URL instead of uploading inline.

Ready to run this?

Launch from the dashboard, CLI, or Compute API. Max-cost protection included.