File size: 2,468 Bytes
835aed0 d8f8753 116fd0c d8f8753 116fd0c a627ca9 116fd0c d8f8753 835aed0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 |
---
license: apache-2.0
---
# Z-Image Turbo Acceleration Capability Fix LoRA
## Model Introduction
This model is a LoRA used to fix the acceleration capability of Z-Image Turbo LoRA.
LoRAs trained directly based on Z-Image Turbo will lose their acceleration capability. Images generated under acceleration configuration (steps=8, cfg=1) become blurry, while images generated under non-acceleration configuration (steps=30, cfg=2) remain normal.
## Results
Training Data:
Generation Results:
|steps=8, cfg=1|steps=30, cfg=2|steps=8, cfg=1, with our model fix|
|-|-|-|
||||
## Training with Z-Image Turbo
If you want to train LoRAs based on Z-Image Turbo while maintaining its acceleration capability, please refer to our detailed training strategies guide:
📖 [**Training Strategies of Z-Image Turbo**](https://huggingface.co/blog/kelseye/training-strategies-of-z-image-turbo)
This guide covers four different training approaches:
- **Scheme 1**: Standard SFT Training + No Acceleration Configuration
- **Scheme 2**: Differential LoRA Training + Acceleration Configuration
- **Scheme 3**: Standard SFT + Trajectory Imitation Distillation + Acceleration Configuration
- **Scheme 4**: Standard SFT + Loading DistillPatch LoRA (Recommended) + Acceleration Configuration
We recommend **Scheme 4** as it offers the best trade-off between training simplicity and inference speed.
## Inference Code
```python
from diffsynth.pipelines.z_image import ZImagePipeline, ModelConfig
import torch
pipe = ZImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="transformer/*.safetensors"),
ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="text_encoder/*.safetensors"),
ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
],
tokenizer_config=ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="tokenizer/"),
)
pipe.load_lora(pipe.dit, "path/to/your/lora.safetensors")
pipe.load_lora(pipe.dit, ModelConfig(model_id="DiffSynth-Studio/Z-Image-Turbo-DistillPatch", origin_file_pattern="model.safetensors"))
image = pipe(prompt="a dog", seed=42, rand_device="cuda")
image.save("image.jpg")
``` |