| --- |
| license: apache-2.0 |
| --- |
| # Templates - Content Reference (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 can extract visual features from an input reference image and fuse them into the base generation guided by natural language descriptions. |
|
|
| * 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 cat is sitting on a stone. |
|
|
| | Template | Generated | Template | Generated | |
| |:---:|:---:|:---:|:---:| |
| |  |  |  |  | |
|
|
| --- |
|
|
| > **Prompt:** A cozy wooden cottage in a lush green valley, white fluffy clouds in the sky, peaceful atmosphere. |
|
|
| | Template | Generated | Template | Generated | |
| |:---:|:---:|:---:|:---:| |
| |  |  |  |  | |
|
|
| --- |
|
|
| > **Prompt:** A beautiful girl on an outdoor adventure. |
|
|
| | Template | Generated | Template | Generated | |
| |:---:|:---:|:---:|:---:| |
| |  |  |  |  | |
|
|
| ## 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 |
| from modelscope import dataset_snapshot_download |
| from PIL import Image |
| import numpy as np |
| ``` |
|
|
| ```python |
| 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/"), |
| ) |
| pipe.dit = pipe.enable_lora_hot_loading(pipe.dit) # Important! |
| template = TemplatePipeline.from_pretrained( |
| torch_dtype=torch.bfloat16, |
| device="cuda", |
| model_configs=[ModelConfig(model_id="DiffSynth-Studio/Template-KleinBase4B-ContentRef")], |
| ) |
| dataset_snapshot_download( |
| "DiffSynth-Studio/examples_in_diffsynth", |
| allow_file_pattern=["templates/*"], |
| local_dir="data/examples", |
| ) |
| image = template( |
| pipe, |
| prompt="A cat is sitting on a stone.", |
| seed=0, cfg_scale=4, num_inference_steps=50, |
| template_inputs=[{ |
| "image": Image.open("data/examples/templates/image_style_1.jpg"), |
| }], |
| negative_template_inputs=[{ |
| "image": Image.fromarray(np.zeros((1024, 1024, 3), dtype=np.uint8) + 128), |
| }], |
| ) |
| image.save("image_ContentRef_1.jpg") |
| image = template( |
| pipe, |
| prompt="A cat is sitting on a stone.", |
| seed=0, cfg_scale=4, num_inference_steps=50, |
| template_inputs=[{ |
| "image": Image.open("data/examples/templates/image_style_2.jpg"), |
| }], |
| negative_template_inputs=[{ |
| "image": Image.fromarray(np.zeros((1024, 1024, 3), dtype=np.uint8) + 128), |
| }], |
| ) |
| image.save("image_ContentRef_2.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 |
| from modelscope import dataset_snapshot_download |
| from PIL import Image |
| import numpy as np |
| ``` |
|
|
| ```python |
| 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-ContentRef")], |
| lazy_loading=True, |
| ) |
| dataset_snapshot_download( |
| "DiffSynth-Studio/examples_in_diffsynth", |
| allow_file_pattern=["templates/*"], |
| local_dir="data/examples", |
| ) |
| image = template( |
| pipe, |
| prompt="A cat is sitting on a stone.", |
| seed=0, cfg_scale=4, num_inference_steps=50, |
| template_inputs=[{ |
| "image": Image.open("data/examples/templates/image_style_1.jpg"), |
| }], |
| negative_template_inputs=[{ |
| "image": Image.fromarray(np.zeros((1024, 1024, 3), dtype=np.uint8) + 128), |
| }], |
| ) |
| image.save("image_ContentRef_1.jpg") |
| image = template( |
| pipe, |
| prompt="A cat is sitting on a stone.", |
| seed=0, cfg_scale=4, num_inference_steps=50, |
| template_inputs=[{ |
| "image": Image.open("data/examples/templates/image_style_2.jpg"), |
| }], |
| negative_template_inputs=[{ |
| "image": Image.fromarray(np.zeros((1024, 1024, 3), dtype=np.uint8) + 128), |
| }], |
| ) |
| image.save("image_ContentRef_2.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-ContentRef/*" --local_dir ./data/diffsynth_example_dataset |
| |
| accelerate launch examples/flux2/model_training/train.py \ |
| --dataset_base_path data/diffsynth_example_dataset/flux2/Template-KleinBase4B-ContentRef \ |
| --dataset_metadata_path data/diffsynth_example_dataset/flux2/Template-KleinBase4B-ContentRef/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-ContentRef:" \ |
| --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-ContentRef_full" \ |
| --trainable_models "template_model" \ |
| --use_gradient_checkpointing \ |
| --find_unused_parameters \ |
| --enable_lora_hot_loading |
| ``` |
