Update README.md

3fb0275 verified 10 days ago

7.89 kB

	---
	license: apache-2.0
	---
	# Templates - Age Control (FLUX.2-klein-base-4B)

	This model is one of the Diffusion Templates series models open-sourced by [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio). It allows direct control over the age of the person in the generated image by inputting the `age` parameter.

	* Open-source code: [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio)
	* Technical report: [arXiv](https://arxiv.org/abs/2604.24351)
	* Project page: [GitHub](https://modelscope.github.io/diffusion-templates-web/)
	* Documentation: [English Version](https://diffsynth-studio-doc.readthedocs.io/en/latest/Diffusion_Templates/Introducing_Diffusion_Templates.html)、[中文版](https://diffsynth-studio-doc.readthedocs.io/zh-cn/latest/Diffusion_Templates/Introducing_Diffusion_Templates.html)
	* Online demo: [ModelScope](https://modelscope.cn/studios/DiffSynth-Studio/Diffusion-Templates)
	* Models: [ModelScope](https://modelscope.cn/collections/DiffSynth-Studio/KleinBase4B-Templates)、[ModelScope International](https://modelscope.ai/collections/DiffSynth-Studio/KleinBase4B-Templates)、[HuggingFace](https://huggingface.co/collections/DiffSynth-Studio/kleinbase4b-templates)
	* Datasets: [ModelScope](https://modelscope.cn/collections/DiffSynth-Studio/ImagePulseV2)、[ModelScope International](https://modelscope.cn/collections/DiffSynth-Studio/ImagePulseV2)、[HuggingFace](https://huggingface.co/collections/DiffSynth-Studio/imagepulsev2)

	## Results

	> Prompt: A portrait of a woman with black hair, wearing a suit.

	\| Age = 20 \| Age = 50 \| Age = 80 \|
	\|:---:\|:---:\|:---:\|
	\| ![](./assets/image1_age_20.jpg) \| ![](./assets/image1_age_50.jpg) \| ![](./assets/image1_age_80.jpg) \|

	---

	> Prompt: A portrait of a man, autumn park background, warm evening sunlight.

	\| Age = 20 \| Age = 50 \| Age = 80 \|
	\|:---:\|:---:\|:---:\|
	\| ![](./assets/image2_age_20.jpg) \| ![](./assets/image2_age_50.jpg) \| ![](./assets/image2_age_80.jpg) \|

	---
	A fashion portrait of an elegant woman wearing a red silk dress, high fashion photography, soft lighting. A modern minimalist living room with furniture.

	\| Age = 20 \| Age = 50 \| Age = 80 \|
	\|:---:\|:---:\|:---:\|
	\| ![](./assets/image3_age_20.jpg) \| ![](./assets/image3_age_50.jpg) \| ![](./assets/image3_age_80.jpg) \|

	## Inference Code

	* Install [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio)

	```
	git clone https://github.com/modelscope/DiffSynth-Studio.git
	cd DiffSynth-Studio
	pip install -e .
	```

	* Direct inference (requires 40G GPU memory)

	```python
	from diffsynth.diffusion.template import TemplatePipeline
	from diffsynth.pipelines.flux2_image import Flux2ImagePipeline, ModelConfig
	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-Age")],
	)
	image = template(
	pipe,
	prompt="A portrait of a woman with black hair, wearing a suit.",
	seed=0, cfg_scale=4, num_inference_steps=50,
	template_inputs=[{"age": 20}],
	negative_template_inputs=[{"age": 45}],
	)
	image.save(f"image_age_20.jpg")
	image = template(
	pipe,
	prompt="A portrait of a woman with black hair, wearing a suit.",
	seed=0, cfg_scale=4, num_inference_steps=50,
	template_inputs=[{"age": 50}],
	negative_template_inputs=[{"age": 45}],
	)
	image.save(f"image_age_50.jpg")
	image = template(
	pipe,
	prompt="A portrait of a woman with black hair, wearing a suit.",
	seed=0, cfg_scale=4, num_inference_steps=50,
	template_inputs=[{"age": 80}],
	negative_template_inputs=[{"age": 45}],
	)
	image.save(f"image_age_80.jpg")
	```

	* Enable lazy loading and memory management, requires 24G GPU memory

	```python
	from diffsynth.diffusion.template import TemplatePipeline
	from diffsynth.pipelines.flux2_image import Flux2ImagePipeline, ModelConfig
	import torch

	vram_config = {
	"offload_dtype": "disk",
	"offload_device": "disk",
	"onload_dtype": torch.float8_e4m3fn,
	"onload_device": "cpu",
	"preparing_dtype": torch.float8_e4m3fn,
	"preparing_device": "cuda",
	"computation_dtype": torch.bfloat16,
	"computation_device": "cuda",
	}
	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", *vram_config),
	ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="text_encoder/.safetensors", *vram_config),
	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/"),
	vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
	)
	template = TemplatePipeline.from_pretrained(
	torch_dtype=torch.bfloat16,
	device="cuda",
	model_configs=[ModelConfig(model_id="DiffSynth-Studio/Template-KleinBase4B-Age")],
	lazy_loading=True,
	)
	image = template(
	pipe,
	prompt="A portrait of a woman with black hair, wearing a suit.",
	seed=0, cfg_scale=4, num_inference_steps=50,
	template_inputs=[{"age": 20}],
	negative_template_inputs=[{"age": 45}],
	)
	image.save(f"image_age_20.jpg")
	image = template(
	pipe,
	prompt="A portrait of a woman with black hair, wearing a suit.",
	seed=0, cfg_scale=4, num_inference_steps=50,
	template_inputs=[{"age": 50}],
	negative_template_inputs=[{"age": 45}],
	)
	image.save(f"image_age_50.jpg")
	image = template(
	pipe,
	prompt="A portrait of a woman with black hair, wearing a suit.",
	seed=0, cfg_scale=4, num_inference_steps=50,
	template_inputs=[{"age": 80}],
	negative_template_inputs=[{"age": 45}],
	)
	image.save(f"image_age_80.jpg")
	```

	## Training Code

	After installing DiffSynth-Studio, use the following script to start training. For more information, please refer to the [DiffSynth-Studio Documentation](https://diffsynth-studio-doc.readthedocs.io/zh-cn/latest/).

	```shell
	modelscope download --dataset DiffSynth-Studio/diffsynth_example_dataset --include "flux2/Template-KleinBase4B-Age/*" --local_dir ./data/diffsynth_example_dataset

	accelerate launch examples/flux2/model_training/train.py \
	--dataset_base_path data/diffsynth_example_dataset/flux2/Template-KleinBase4B-Age \
	--dataset_metadata_path data/diffsynth_example_dataset/flux2/Template-KleinBase4B-Age/metadata.jsonl \
	--extra_inputs "template_inputs" \
	--max_pixels 1048576 \
	--dataset_repeat 50 \
	--model_id_with_origin_paths "black-forest-labs/FLUX.2-klein-4B:text_encoder/.safetensors,black-forest-labs/FLUX.2-klein-base-4B:transformer/.safetensors,black-forest-labs/FLUX.2-klein-4B:vae/diffusion_pytorch_model.safetensors" \
	--template_model_id_or_path "DiffSynth-Studio/Template-KleinBase4B-Age:" \
	--tokenizer_path "black-forest-labs/FLUX.2-klein-4B:tokenizer/" \
	--learning_rate 1e-4 \
	--num_epochs 2 \
	--remove_prefix_in_ckpt "pipe.template_model." \
	--output_path "./models/train/Template-KleinBase4B-Age_full" \
	--trainable_models "template_model" \
	--use_gradient_checkpointing \
	--find_unused_parameters
	```