theluyuan/DiffSynth-Studio
1
0
Fork 0
mirror of https://github.com/modelscope/DiffSynth-Studio.git synced 2026-03-18 22:08:13 +00:00
Code Issues Packages Projects Releases Wiki Activity

support z-image-omni-base training

Browse Source
This commit is contained in:
Artiprocher
2026-01-05 20:04:00 +08:00
parent 5745c9f200
commit 32449a6aa0
9 changed files with 128 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
examples/z_image/model_training/full/Z-Image-Omni-Base.sh Normal file
View File

@@ -0,0 +1,14 @@
# This example is tested on 8*A100
accelerate launch --config_file examples/z_image/model_training/full/accelerate_config.yaml examples/z_image/model_training/train.py \
--dataset_base_path data/example_image_dataset \
--dataset_metadata_path data/example_image_dataset/metadata.csv \
--max_pixels 1048576 \
--dataset_repeat 400 \
--model_id_with_origin_paths "Tongyi-MAI/Z-Image-Omni-Base:transformer/*.safetensors,Tongyi-MAI/Z-Image-Omni-Base:siglip/model.safetensors,Tongyi-MAI/Z-Image-Turbo:text_encoder/*.safetensors,Tongyi-MAI/Z-Image-Turbo:vae/diffusion_pytorch_model.safetensors" \
--learning_rate 1e-5 \
--num_epochs 2 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/Z-Image-Omni-Base_full" \
--trainable_models "dit" \
--use_gradient_checkpointing \
--dataset_num_workers 8

15
examples/z_image/model_training/lora/Z-Image-Omni-Base.sh Normal file
View File

@@ -0,0 +1,15 @@
accelerate launch examples/z_image/model_training/train.py \
--dataset_base_path data/example_image_dataset \
--dataset_metadata_path data/example_image_dataset/metadata.csv \
--max_pixels 1048576 \
--dataset_repeat 50 \
--model_id_with_origin_paths "Tongyi-MAI/Z-Image-Omni-Base:transformer/*.safetensors,Tongyi-MAI/Z-Image-Omni-Base:siglip/model.safetensors,Tongyi-MAI/Z-Image-Turbo:text_encoder/*.safetensors,Tongyi-MAI/Z-Image-Turbo:vae/diffusion_pytorch_model.safetensors" \
--learning_rate 1e-4 \
--num_epochs 5 \
--remove_prefix_in_ckpt "pipe.dit." \
--output_path "./models/train/Z-Image-Omni-Base_lora" \
--lora_base_model "dit" \
--lora_target_modules "to_q,to_k,to_v,to_out.0,w1,w2,w3" \
--lora_rank 32 \
--use_gradient_checkpointing \
--dataset_num_workers 8

21
examples/z_image/model_training/validate_full/Z-Image-Omni-Base.py Normal file
View File

@@ -0,0 +1,21 @@
from diffsynth.pipelines.z_image import ZImagePipeline, ModelConfig
from diffsynth.core import load_state_dict
import torch
pipe = ZImagePipeline.from_pretrained(
torch_dtype=torch.bfloat16,
device="cuda",
model_configs=[
ModelConfig(model_id="Tongyi-MAI/Z-Image-Omni-Base", origin_file_pattern="transformer/*.safetensors"),
ModelConfig(model_id="Tongyi-MAI/Z-Image-Omni-Base", origin_file_pattern="siglip/model.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/"),
)
state_dict = load_state_dict("./models/train/Z-Image-Omni-Base_full/epoch-1.safetensors", torch_dtype=torch.bfloat16)
pipe.dit.load_state_dict(state_dict)
prompt = "a dog"
image = pipe(prompt=prompt, seed=42, rand_device="cuda", num_inference_steps=40, cfg_scale=4)
image.save("image.jpg")

19
examples/z_image/model_training/validate_lora/Z-Image-Omni-Base.py Normal file
View File

@@ -0,0 +1,19 @@
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-Omni-Base", origin_file_pattern="transformer/*.safetensors"),
ModelConfig(model_id="Tongyi-MAI/Z-Image-Omni-Base", origin_file_pattern="siglip/model.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, "./models/train/Z-Image-Omni-Base_lora/epoch-4.safetensors")
prompt = "a dog"
image = pipe(prompt=prompt, seed=42, rand_device="cuda", num_inference_steps=40, cfg_scale=4)
image.save("image.jpg")