import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Comfy-Org/z_image", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("fwwrsd/tvix-epoch20")
prompt = "tvix"
image = pipe(prompt).images[0]tvix — Epoch 20/50
Z_IMAGE_BASE identity LoRA checkpoint (epoch 20 of 50).
Trigger Word
Use tvix in your prompt.
Download
https://huggingface.co/fwwrsd/tvix-epoch20/resolve/main/tvix_epoch20.safetensors
Training Details
| Parameter | Value |
|---|---|
| Base Model | Comfy-Org/z_image |
| Epoch | 20 / 50 |
| LoRA Rank | 32 |
| Learning Rate | 0.00008 |
| Resolution | 1024px |
| Training Media | 61 |
Trained with NanoBanana LoRA Bot on RunPod
- Downloads last month
- -
Model tree for fwwrsd/tvix-epoch20
Base model
Comfy-Org/z_image