Merge branch 'main' into lora-hot-swapping

Author: BenjaminBossan

docs/source/en/api/pipelines/lumina2.md

@@ -26,6 +26,56 @@ Make sure to check out the Schedulers [guide](../../using-diffusers/schedulers)
 
 </Tip>
 
+## Using Single File loading with Lumina Image 2.0
+
+Single file loading for Lumina Image 2.0 is available for the `Lumina2Transformer2DModel`
+
+```python
+import torch
+from diffusers import Lumina2Transformer2DModel, Lumina2Text2ImgPipeline
+
+ckpt_path = "https://huggingface.co/Alpha-VLLM/Lumina-Image-2.0/blob/main/consolidated.00-of-01.pth"
+transformer = Lumina2Transformer2DModel.from_single_file(
+    ckpt_path, torch_dtype=torch.bfloat16
+)
+
+pipe = Lumina2Text2ImgPipeline.from_pretrained(
+    "Alpha-VLLM/Lumina-Image-2.0", transformer=transformer, torch_dtype=torch.bfloat16
+)
+pipe.enable_model_cpu_offload()
+image = pipe(
+    "a cat holding a sign that says hello",
+    generator=torch.Generator("cpu").manual_seed(0),
+).images[0]
+image.save("lumina-single-file.png")
+
+```
+
+## Using GGUF Quantized Checkpoints with Lumina Image 2.0
+
+GGUF Quantized checkpoints for the `Lumina2Transformer2DModel` can be loaded via `from_single_file` with the `GGUFQuantizationConfig` 
+
+```python
+from diffusers import Lumina2Transformer2DModel, Lumina2Text2ImgPipeline, GGUFQuantizationConfig 
+
+ckpt_path = "https://huggingface.co/calcuis/lumina-gguf/blob/main/lumina2-q4_0.gguf"
+transformer = Lumina2Transformer2DModel.from_single_file(
+    ckpt_path,
+    quantization_config=GGUFQuantizationConfig(compute_dtype=torch.bfloat16),
+    torch_dtype=torch.bfloat16,
+)
+
+pipe = Lumina2Text2ImgPipeline.from_pretrained(
+    "Alpha-VLLM/Lumina-Image-2.0", transformer=transformer, torch_dtype=torch.bfloat16
+)
+pipe.enable_model_cpu_offload()
+image = pipe(
+    "a cat holding a sign that says hello",
+    generator=torch.Generator("cpu").manual_seed(0),
+).images[0]
+image.save("lumina-gguf.png")
+```
+
 ## Lumina2Text2ImgPipeline
 
 [[autodoc]] Lumina2Text2ImgPipeline

docs/source/en/api/utilities.md

@@ -45,3 +45,7 @@ Utility and helper functions for working with 🤗 Diffusers.
 ## apply_layerwise_casting
 
 [[autodoc]] hooks.layerwise_casting.apply_layerwise_casting
+
+## apply_group_offloading
+
+[[autodoc]] hooks.group_offloading.apply_group_offloading

docs/source/en/optimization/memory.md

@@ -158,6 +158,46 @@ In order to properly offload models after they're called, it is required to run
 
 </Tip>
 
+## Group offloading
+
+Group offloading is the middle ground between sequential and model offloading. It works by offloading groups of internal layers (either `torch.nn.ModuleList` or `torch.nn.Sequential`), which uses less memory than model-level offloading. It is also faster than sequential-level offloading because the number of device synchronizations is reduced.
+
+To enable group offloading, call the [`~ModelMixin.enable_group_offload`] method on the model if it is a Diffusers model implementation. For any other model implementation, use [`~hooks.group_offloading.apply_group_offloading`]:
+
+```python
+import torch
+from diffusers import CogVideoXPipeline
+from diffusers.hooks import apply_group_offloading
+from diffusers.utils import export_to_video
+
+# Load the pipeline
+onload_device = torch.device("cuda")
+offload_device = torch.device("cpu")
+pipe = CogVideoXPipeline.from_pretrained("THUDM/CogVideoX-5b", torch_dtype=torch.bfloat16)
+
+# We can utilize the enable_group_offload method for Diffusers model implementations
+pipe.transformer.enable_group_offload(onload_device=onload_device, offload_device=offload_device, offload_type="leaf_level", use_stream=True)
+
+# For any other model implementations, the apply_group_offloading function can be used
+apply_group_offloading(pipe.text_encoder, onload_device=onload_device, offload_type="block_level", num_blocks_per_group=2)
+apply_group_offloading(pipe.vae, onload_device=onload_device, offload_type="leaf_level")
+
+prompt = (
+    "A panda, dressed in a small, red jacket and a tiny hat, sits on a wooden stool in a serene bamboo forest. "
+    "The panda's fluffy paws strum a miniature acoustic guitar, producing soft, melodic tunes. Nearby, a few other "
+    "pandas gather, watching curiously and some clapping in rhythm. Sunlight filters through the tall bamboo, "
+    "casting a gentle glow on the scene. The panda's face is expressive, showing concentration and joy as it plays. "
+    "The background includes a small, flowing stream and vibrant green foliage, enhancing the peaceful and magical "
+    "atmosphere of this unique musical performance."
+)
+video = pipe(prompt=prompt, guidance_scale=6, num_inference_steps=50).frames[0]
+# This utilized about 14.79 GB. It can be further reduced by using tiling and using leaf_level offloading throughout the pipeline.
+print(f"Max memory reserved: {torch.cuda.max_memory_allocated() / 1024**3:.2f} GB")
+export_to_video(video, "output.mp4", fps=8)
+```
+
+Group offloading (for CUDA devices with support for asynchronous data transfer streams) overlaps data transfer and computation to reduce the overall execution time compared to sequential offloading. This is enabled using layer prefetching with CUDA streams. The next layer to be executed is loaded onto the accelerator device while the current layer is being executed - this increases the memory requirements slightly. Group offloading also supports leaf-level offloading (equivalent to sequential CPU offloading) but can be made much faster when using streams.
+
 ## FP8 layerwise weight-casting
 
 PyTorch supports `torch.float8_e4m3fn` and `torch.float8_e5m2` as weight storage dtypes, but they can't be used for computation in many different tensor operations due to unimplemented kernel support. However, you can use these dtypes to store model weights in fp8 precision and upcast them on-the-fly when the layers are used in the forward pass. This is known as layerwise weight-casting.

docs/source/en/tutorials/using_peft_for_inference.md

@@ -221,3 +221,7 @@ pipe.delete_adapters("toy")
 pipe.get_active_adapters()
 ["pixel"]
 ```
+
+## PeftInputAutocastDisableHook
+
+[[autodoc]] hooks.layerwise_casting.PeftInputAutocastDisableHook

examples/community/README.md

@@ -50,8 +50,9 @@ Please also check out our [Community Scripts](https://github.com/huggingface/dif
 |   IADB Pipeline                                                                                                    | Implementation of [Iterative α-(de)Blending: a Minimalist Deterministic Diffusion Model](https://arxiv.org/abs/2305.03486)                                                                                                                                                                                                                                                                                                                                                                                                                                      | [IADB Pipeline](#iadb-pipeline)      | - |              [Thomas Chambon](https://github.com/tchambon)
 |   Zero1to3 Pipeline                                                                                                    | Implementation of [Zero-1-to-3: Zero-shot One Image to 3D Object](https://arxiv.org/abs/2303.11328)                                                                                                                                                                                                                                                                                                                                                                                                                                      | [Zero1to3 Pipeline](#zero1to3-pipeline)      | - |              [Xin Kong](https://github.com/kxhit) |
 | Stable Diffusion XL Long Weighted Prompt Pipeline | A pipeline support unlimited length of prompt and negative prompt, use A1111 style of prompt weighting | [Stable Diffusion XL Long Weighted Prompt Pipeline](#stable-diffusion-xl-long-weighted-prompt-pipeline) | [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1LsqilswLR40XLLcp6XFOl5nKb_wOe26W?usp=sharing) | [Andrew Zhu](https://xhinker.medium.com/) |
-| Stable Diffusion Mixture Tiling Pipeline SD 1.5 | A pipeline generates cohesive images by integrating multiple diffusion processes, each focused on a specific image region and considering boundary effects for smooth blending | [Stable Diffusion Mixture Tiling Pipeline SD 1.5](#stable-diffusion-mixture-tiling-sd-15) | [![Hugging Face Space](https://img.shields.io/badge/🤗%20Hugging%20Face-Space-yellow)](https://huggingface.co/spaces/albarji/mixture-of-diffusers) | [Álvaro B Jiménez](https://github.com/albarji/) |
-| Stable Diffusion Mixture Tiling Pipeline SDXL | A pipeline generates cohesive images by integrating multiple diffusion processes, each focused on a specific image region and considering boundary effects for smooth blending | [Stable Diffusion Mixture Tiling Pipeline SDXL](#stable-diffusion-mixture-tiling-sdxl) | [![Hugging Face Space](https://img.shields.io/badge/🤗%20Hugging%20Face-Space-yellow)](https://huggingface.co/spaces/elismasilva/mixture-of-diffusers-sdxl-tiling) | [Eliseu Silva](https://github.com/DEVAIEXP/) |
+| Stable Diffusion Mixture Tiling Pipeline SD 1.5 | A pipeline generates cohesive images by integrating multiple diffusion processes, each focused on a specific image region and considering boundary effects for smooth blending | [Stable Diffusion Mixture Tiling Pipeline SD 1.5](#stable-diffusion-mixture-tiling-pipeline-sd-15) | [![Hugging Face Space](https://img.shields.io/badge/🤗%20Hugging%20Face-Space-yellow)](https://huggingface.co/spaces/albarji/mixture-of-diffusers) | [Álvaro B Jiménez](https://github.com/albarji/) |
+| Stable Diffusion Mixture Canvas Pipeline SD 1.5 | A pipeline generates cohesive images by integrating multiple diffusion processes, each focused on a specific image region and considering boundary effects for smooth blending. Works by defining a list of Text2Image region objects that detail the region of influence of each diffuser. | [Stable Diffusion Mixture Canvas Pipeline SD 1.5](#stable-diffusion-mixture-canvas-pipeline-sd-15) | [![Hugging Face Space](https://img.shields.io/badge/🤗%20Hugging%20Face-Space-yellow)](https://huggingface.co/spaces/albarji/mixture-of-diffusers) | [Álvaro B Jiménez](https://github.com/albarji/) |
+| Stable Diffusion Mixture Tiling Pipeline SDXL | A pipeline generates cohesive images by integrating multiple diffusion processes, each focused on a specific image region and considering boundary effects for smooth blending | [Stable Diffusion Mixture Tiling Pipeline SDXL](#stable-diffusion-mixture-tiling-pipeline-sdxl) | [![Hugging Face Space](https://img.shields.io/badge/🤗%20Hugging%20Face-Space-yellow)](https://huggingface.co/spaces/elismasilva/mixture-of-diffusers-sdxl-tiling) | [Eliseu Silva](https://github.com/DEVAIEXP/) |
 | FABRIC - Stable Diffusion with feedback Pipeline | pipeline supports feedback from liked and disliked images | [Stable Diffusion Fabric Pipeline](#stable-diffusion-fabric-pipeline) | [Notebook](https://github.com/huggingface/notebooks/blob/main/diffusers/stable_diffusion_fabric.ipynb)| [Shauray Singh](https://shauray8.github.io/about_shauray/) |
 | sketch inpaint - Inpainting with non-inpaint Stable Diffusion | sketch inpaint much like in automatic1111 | [Masked Im2Im Stable Diffusion Pipeline](#stable-diffusion-masked-im2im) | - | [Anatoly Belikov](https://github.com/noskill) |
 | sketch inpaint xl - Inpainting with non-inpaint Stable Diffusion | sketch inpaint much like in automatic1111 | [Masked Im2Im Stable Diffusion XL Pipeline](#stable-diffusion-xl-masked-im2im) | - | [Anatoly Belikov](https://github.com/noskill) |
@@ -2404,7 +2405,7 @@ pipe_images = mixing_pipeline(
 
 ![image_mixing_result](https://huggingface.co/datasets/TheDenk/images_mixing/resolve/main/boromir_gigachad.png)
 
-### Stable Diffusion Mixture Tiling SD 1.5
+### Stable Diffusion Mixture Tiling Pipeline SD 1.5
 
 This pipeline uses the Mixture. Refer to the [Mixture](https://arxiv.org/abs/2302.02412) paper for more details.
 
@@ -2435,7 +2436,7 @@ image = pipeline(
 
 ![mixture_tiling_results](https://huggingface.co/datasets/kadirnar/diffusers_readme_images/resolve/main/mixture_tiling.png)
 
-### Stable Diffusion Mixture Canvas
+### Stable Diffusion Mixture Canvas Pipeline SD 1.5
 
 This pipeline uses the Mixture. Refer to the [Mixture](https://arxiv.org/abs/2302.02412) paper for more details.
 
@@ -2470,7 +2471,7 @@ output = pipeline(
 ![Input_Image](https://huggingface.co/datasets/kadirnar/diffusers_readme_images/resolve/main/input_image.png)
 ![mixture_canvas_results](https://huggingface.co/datasets/kadirnar/diffusers_readme_images/resolve/main/canvas.png)
 
-### Stable Diffusion Mixture Tiling SDXL
+### Stable Diffusion Mixture Tiling Pipeline SDXL
 
 This pipeline uses the Mixture. Refer to the [Mixture](https://arxiv.org/abs/2302.02412) paper for more details.
 
@@ -2516,14 +2517,13 @@ image = pipe(
     tile_col_overlap=256,
     guidance_scale_tiles=[[7, 7, 7]], # or guidance_scale=7 if is the same for all prompts
     height=1024,
-    width=3840,
-    target_size=(1024, 3840),
+    width=3840,    
     generator=generator,
     num_inference_steps=30,
 )["images"][0]
 ```
 
-![mixture_tiling_results](https://huggingface.co/datasets/elismasilva/results/resolve/main/mixture_sdxl.png)
+![mixture_tiling_results](https://huggingface.co/datasets/elismasilva/results/resolve/main/mixture_of_diffusers_sdxl_1.png)
 
 ### TensorRT Inpainting Stable Diffusion Pipeline