DiffSynth-Studio/examples/flux2/model_training/validate_full/Template-KleinBase4B-SoftRGB.py at main · modelscope/DiffSynth-Studio · GitHub

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
from diffsynth.diffusion.template import TemplatePipeline
from diffsynth.pipelines.flux2_image import Flux2ImagePipeline, ModelConfig
from diffsynth.core import load_state_dict
import torch

pipe = Flux2ImagePipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[
        ModelConfig(model_id="black-forest-labs/FLUX.2-klein-base-4B", origin_file_pattern="transformer/*.safetensors"),
        ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="text_encoder/*.safetensors"),
        ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
    ],
    tokenizer_config=ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="tokenizer/"),
)
template = TemplatePipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[ModelConfig(model_id="DiffSynth-Studio/Template-KleinBase4B-SoftRGB")],
)
state_dict = load_state_dict("./models/train/Template-KleinBase4B-SoftRGB_full/epoch-1.safetensors", torch_dtype=torch.bfloat16)
template.models[0].load_state_dict(state_dict)
image = template(
    pipe,
    prompt="A cat is sitting on a stone.",
    seed=0, cfg_scale=4, num_inference_steps=50,
    template_inputs = [{
        "R": 128/255,
        "G": 128/255,
        "B": 128/255
    }],
)
image.save("image_rgb_normal.jpg")
image = template(
    pipe,
    prompt="A cat is sitting on a stone.",
    seed=0, cfg_scale=4, num_inference_steps=50,
    template_inputs = [{
        "R": 208/255,
        "G": 185/255,
        "B": 138/255
    }],
)
image.save("image_rgb_warm.jpg")
image = template(
    pipe,
    prompt="A cat is sitting on a stone.",
    seed=0, cfg_scale=4, num_inference_steps=50,
    template_inputs = [{
        "R": 94/255,
        "G": 163/255,
        "B": 174/255
    }],
)
image.save("image_rgb_cold.jpg")