| # cl4ud1a β LoRA adapter (Z-Image Turbo, rank=16) | |
| This folder holds the training experiment named `cl4ud1a` β a LoRA adapter for Tongyi-MAI/Z-Image-Turbo (Z-Image Turbo) tuned with rank 16 parameters. | |
| Key facts | |
| - Base model: Tongyi-MAI/Z-Image-Turbo (arch: zimage:turbo) | |
| - LoRA type: UNet LoRA (linear=16, conv=16, alpha=16) | |
| - Training steps: 3000 (checkpoints saved every 250 steps) | |
| - Save format: Diffusers safetensors (dtype: bf16) | |
| - Training device: cuda | |
| - Quantization: qfloat8 applied to model and text encoder | |
| Artifacts in this folder | |
| - `cl4ud1a.safetensors` β final LoRA/adapted weights (merged in training pipeline) | |
| - `cl4ud1a_00000XXXXX.safetensors` β saved checkpoints (step increments) | |
| - `optimizer.pt` β optimizer state (checkpoint) | |
| - `config.yaml` β original run configuration used for training | |
| - `log.txt` β raw training log (progress, warnings, reproducibility notes) | |
| - `samples/` β generated sample images at a few checkpoints | |
| Training & notes observed | |
| - Training used a small dataset (~14 images, augmented to 28 via flips) at mixed resolutions (768β1024). Latents and text embeddings were cached for speed. | |
| - A PIL-based EXIF parsing error appeared for one PNG during preprocessing; dataset sanitation is recommended before reproduction (see log snippet). | |
| - Assistant LoRA adapter was loaded/merged during training β see `config.yaml` for assistant adapter path. | |
| How to reproduce (short) | |
| 1. Ensure you have the same base model (Tongyi-MAI/Z-Image-Turbo) accessible. | |
| 2. Recreate the environment with GPU + CUDA and BF16 support. | |
| 3. Use `config.yaml` to re-run the trainer used by the author (dataset paths will need adjustment). | |
| Usage example (consumer) | |
| - To apply the LoRA at inference time, use your Z-Image-Turbo-compatible pipeline loader and merge or inject the safetensors file into the UNet weights (example depends on your runner/adapter). | |
| If you plan to upload this experiment to Hugging Face: include `cl4ud1a.safetensors`, `config.yaml`, `log.txt` and a short model card describing license and data provenance. | |