[docs] PyTorch 2.0 (#11618)

* combine

* Update docs/source/en/optimization/fp16.md
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

docs/source/en/_toctree.yml

@@ -180,8 +180,6 @@
     title: Accelerate inference
   - local: optimization/memory
     title: Reduce memory usage
-  - local: optimization/torch2.0
-    title: PyTorch 2.0
   - local: optimization/xformers
     title: xFormers
   - local: optimization/tome

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

@@ -347,7 +347,7 @@ pipe.to("cuda")
 image = pipe(image=image, prompt="<prompt>", strength=0.3).images
 ```
 
-You can also use [`torch.compile`](../../optimization/torch2.0). Note that we have not exhaustively tested `torch.compile`
+You can also use [`torch.compile`](../../optimization/fp16#torchcompile). Note that we have not exhaustively tested `torch.compile`
 with IF and it might not give expected results.
 
 ```py

docs/source/en/optimization/fp16.md

@@ -150,6 +150,24 @@ pipeline(prompt, num_inference_steps=30).images[0]
 
 Compilation is slow the first time, but once compiled, it is significantly faster. Try to only use the compiled pipeline on the same type of inference operations. Calling the compiled pipeline on a different image size retriggers compilation which is slow and inefficient.
 
+### Regional compilation
+
+[Regional compilation](https://docs.pytorch.org/tutorials/recipes/regional_compilation.html) reduces the cold start compilation time by only compiling a specific repeated region (or block) of the model instead of the entire model. The compiler reuses the cached and compiled code for the other blocks.
+
+[Accelerate](https://huggingface.co/docs/accelerate/index) provides the [compile_regions](https://github.com/huggingface/accelerate/blob/273799c85d849a1954a4f2e65767216eb37fa089/src/accelerate/utils/other.py#L78) method for automatically compiling the repeated blocks of a `nn.Module` sequentially. The rest of the model is compiled separately.
+
+```py
+# pip install -U accelerate
+import torch
+from diffusers import StableDiffusionXLPipeline
+from accelerate.utils import compile regions
+
+pipeline = StableDiffusionXLPipeline.from_pretrained(
+    "stabilityai/stable-diffusion-xl-base-1.0", torch_dtype=torch.float16
+).to("cuda")
+pipeline.unet = compile_regions(pipeline.unet, mode="reduce-overhead", fullgraph=True)
+```
+
 ### Graph breaks
 
 It is important to specify `fullgraph=True` in torch.compile to ensure there are no graph breaks in the underlying model. This allows you to take advantage of torch.compile without any performance degradation. For the UNet and VAE, this changes how you access the return variables.
@@ -170,6 +188,12 @@ The `step()` function is [called](https://github.com/huggingface/diffusers/blob/
 
 In general, the `sigmas` should [stay on the CPU](https://github.com/huggingface/diffusers/blob/35a969d297cba69110d175ee79c59312b9f49e1e/src/diffusers/schedulers/scheduling_euler_discrete.py#L240) to avoid the communication sync and latency.
 
+### Benchmarks
+
+Refer to the [diffusers/benchmarks](https://huggingface.co/datasets/diffusers/benchmarks) dataset to see inference latency and memory usage data for compiled pipelines.
+
+The [diffusers-torchao](https://github.com/sayakpaul/diffusers-torchao#benchmarking-results) repository also contains benchmarking results for compiled versions of Flux and CogVideoX.
+
 ## Dynamic quantization
 
 [Dynamic quantization](https://pytorch.org/tutorials/recipes/recipes/dynamic_quantization.html) improves inference speed by reducing precision to enable faster math operations. This particular type of quantization determines how to scale the activations based on the data at runtime rather than using a fixed scaling factor. As a result, the scaling factor is more accurately aligned with the data.

docs/source/en/optimization/tome.md

@@ -93,4 +93,4 @@ To reproduce this benchmark, feel free to use this [script](https://gist.github.
 |          |                |              2 |         OOM |             13 |               10.78 |
 |          |                |              1 |         OOM |           6.66 |                5.54 |
 
-As seen in the tables above, the speed-up from `tomesd` becomes more pronounced for larger image resolutions. It is also interesting to note that with `tomesd`, it is possible to run the pipeline on a higher resolution like 1024x1024. You may be able to speed-up inference even more with [`torch.compile`](torch2.0).
+As seen in the tables above, the speed-up from `tomesd` becomes more pronounced for larger image resolutions. It is also interesting to note that with `tomesd`, it is possible to run the pipeline on a higher resolution like 1024x1024. You may be able to speed-up inference even more with [`torch.compile`](fp16#torchcompile).

docs/source/en/quantization/torchao.md

@@ -56,7 +56,7 @@ image = pipe(
 image.save("output.png")
 ```
 
-TorchAO is fully compatible with [torch.compile](./optimization/torch2.0#torchcompile), setting it apart from other quantization methods. This makes it easy to speed up inference with just one line of code.
+TorchAO is fully compatible with [torch.compile](../optimization/fp16#torchcompile), setting it apart from other quantization methods. This makes it easy to speed up inference with just one line of code.
 
 ```python
 # In the above code, add the following after initializing the transformer

docs/source/en/stable_diffusion.md

@@ -256,6 +256,6 @@ make_image_grid(images, 2, 2)
 
 In this tutorial, you learned how to optimize a [`DiffusionPipeline`] for computational and memory efficiency as well as improving the quality of generated outputs. If you're interested in making your pipeline even faster, take a look at the following resources:
 
-- Learn how [PyTorch 2.0](./optimization/torch2.0) and [`torch.compile`](https://pytorch.org/docs/stable/generated/torch.compile.html) can yield 5 - 300% faster inference speed. On an A100 GPU, inference can be up to 50% faster!
+- Learn how [PyTorch 2.0](./optimization/fp16) and [`torch.compile`](https://pytorch.org/docs/stable/generated/torch.compile.html) can yield 5 - 300% faster inference speed. On an A100 GPU, inference can be up to 50% faster!
 - If you can't use PyTorch 2, we recommend you install [xFormers](./optimization/xformers). Its memory-efficient attention mechanism works great with PyTorch 1.13.1 for faster speed and reduced memory consumption.
 - Other optimization techniques, such as model offloading, are covered in [this guide](./optimization/fp16).

docs/source/en/training/overview.md

@@ -59,5 +59,5 @@ pip install -r requirements_sdxl.txt
 
 To speedup training and reduce memory-usage, we recommend:
 
-- using PyTorch 2.0 or higher to automatically use [scaled dot product attention](../optimization/torch2.0#scaled-dot-product-attention) during training (you don't need to make any changes to the training code)
+- using PyTorch 2.0 or higher to automatically use [scaled dot product attention](../optimization/fp16#scaled-dot-product-attention) during training (you don't need to make any changes to the training code)
 - installing [xFormers](../optimization/xformers) to enable memory-efficient attention
\ No newline at end of file

docs/source/en/tutorials/using_peft_for_inference.md

@@ -103,7 +103,7 @@ pipeline("A cute cnmt eating a slice of pizza, stunning color scheme, masterpiec
 
 ## torch.compile
 
-[torch.compile](../optimization/torch2.0#torchcompile) speeds up inference by compiling the PyTorch model to use optimized kernels. Before compiling, the LoRA weights need to be fused into the base model and unloaded first.
+[torch.compile](../optimization/fp16#torchcompile) speeds up inference by compiling the PyTorch model to use optimized kernels. Before compiling, the LoRA weights need to be fused into the base model and unloaded first.
 
 ```py
 import torch

docs/source/en/using-diffusers/conditional_image_generation.md

@@ -303,7 +303,7 @@ There are many types of conditioning inputs you can use, and 🤗 Diffusers supp
 
 Diffusion models are large, and the iterative nature of denoising an image is computationally expensive and intensive. But this doesn't mean you need access to powerful - or even many - GPUs to use them. There are many optimization techniques for running diffusion models on consumer and free-tier resources. For example, you can load model weights in half-precision to save GPU memory and increase speed or offload the entire model to the GPU to save even more memory.
 
-PyTorch 2.0 also supports a more memory-efficient attention mechanism called [*scaled dot product attention*](../optimization/torch2.0#scaled-dot-product-attention) that is automatically enabled if you're using PyTorch 2.0. You can combine this with [`torch.compile`](https://pytorch.org/tutorials/intermediate/torch_compile_tutorial.html) to speed your code up even more:
+PyTorch 2.0 also supports a more memory-efficient attention mechanism called [*scaled dot product attention*](../optimization/fp16#scaled-dot-product-attention) that is automatically enabled if you're using PyTorch 2.0. You can combine this with [`torch.compile`](https://pytorch.org/tutorials/intermediate/torch_compile_tutorial.html) to speed your code up even more:
 
 ```py
 from diffusers import AutoPipelineForText2Image
@@ -313,4 +313,4 @@ pipeline = AutoPipelineForText2Image.from_pretrained("stable-diffusion-v1-5/stab
 pipeline.unet = torch.compile(pipeline.unet, mode="reduce-overhead", fullgraph=True)
 ```
 
-For more tips on how to optimize your code to save memory and speed up inference, read the [Memory and speed](../optimization/fp16) and [Torch 2.0](../optimization/torch2.0) guides.
+For more tips on how to optimize your code to save memory and speed up inference, read the [Accelerate inference](../optimization/fp16) and [Reduce memory usage](../optimization/memory) guides.

docs/source/en/using-diffusers/img2img.md

@@ -35,7 +35,7 @@ pipeline.enable_xformers_memory_efficient_attention()
 
 <Tip>
 
-You'll notice throughout the guide, we use [`~DiffusionPipeline.enable_model_cpu_offload`] and [`~DiffusionPipeline.enable_xformers_memory_efficient_attention`], to save memory and increase inference speed. If you're using PyTorch 2.0, then you don't need to call [`~DiffusionPipeline.enable_xformers_memory_efficient_attention`] on your pipeline because it'll already be using PyTorch 2.0's native [scaled-dot product attention](../optimization/torch2.0#scaled-dot-product-attention).
+You'll notice throughout the guide, we use [`~DiffusionPipeline.enable_model_cpu_offload`] and [`~DiffusionPipeline.enable_xformers_memory_efficient_attention`], to save memory and increase inference speed. If you're using PyTorch 2.0, then you don't need to call [`~DiffusionPipeline.enable_xformers_memory_efficient_attention`] on your pipeline because it'll already be using PyTorch 2.0's native [scaled-dot product attention](../optimization/fp16#scaled-dot-product-attention).
 
 </Tip>
 
@@ -589,17 +589,17 @@ make_image_grid([init_image, depth_image, image_control_net, image_elden_ring],
 
 ## Optimize
 
-Running diffusion models is computationally expensive and intensive, but with a few optimization tricks, it is entirely possible to run them on consumer and free-tier GPUs. For example, you can use a more memory-efficient form of attention such as PyTorch 2.0's [scaled-dot product attention](../optimization/torch2.0#scaled-dot-product-attention) or [xFormers](../optimization/xformers) (you can use one or the other, but there's no need to use both). You can also offload the model to the GPU while the other pipeline components wait on the CPU.
+Running diffusion models is computationally expensive and intensive, but with a few optimization tricks, it is entirely possible to run them on consumer and free-tier GPUs. For example, you can use a more memory-efficient form of attention such as PyTorch 2.0's [scaled-dot product attention](../optimization/fp16#scaled-dot-product-attention) or [xFormers](../optimization/xformers) (you can use one or the other, but there's no need to use both). You can also offload the model to the GPU while the other pipeline components wait on the CPU.
 
 ```diff
 + pipeline.enable_model_cpu_offload()
 + pipeline.enable_xformers_memory_efficient_attention()
 ```
 
-With [`torch.compile`](../optimization/torch2.0#torchcompile), you can boost your inference speed even more by wrapping your UNet with it:
+With [`torch.compile`](../optimization/fp16#torchcompile), you can boost your inference speed even more by wrapping your UNet with it:
 
 ```py
 pipeline.unet = torch.compile(pipeline.unet, mode="reduce-overhead", fullgraph=True)
 ```
 
-To learn more, take a look at the [Reduce memory usage](../optimization/memory) and [Torch 2.0](../optimization/torch2.0) guides.
+To learn more, take a look at the [Reduce memory usage](../optimization/memory) and [Accelerate inference](../optimization/fp16) guides.

docs/source/en/using-diffusers/inpaint.md

@@ -35,7 +35,7 @@ pipeline.enable_xformers_memory_efficient_attention()
 
 <Tip>
 
-You'll notice throughout the guide, we use [`~DiffusionPipeline.enable_model_cpu_offload`] and [`~DiffusionPipeline.enable_xformers_memory_efficient_attention`], to save memory and increase inference speed. If you're using PyTorch 2.0, it's not necessary to call [`~DiffusionPipeline.enable_xformers_memory_efficient_attention`] on your pipeline because it'll already be using PyTorch 2.0's native [scaled-dot product attention](../optimization/torch2.0#scaled-dot-product-attention).
+You'll notice throughout the guide, we use [`~DiffusionPipeline.enable_model_cpu_offload`] and [`~DiffusionPipeline.enable_xformers_memory_efficient_attention`], to save memory and increase inference speed. If you're using PyTorch 2.0, it's not necessary to call [`~DiffusionPipeline.enable_xformers_memory_efficient_attention`] on your pipeline because it'll already be using PyTorch 2.0's native [scaled-dot product attention](../optimization/fp16#scaled-dot-product-attention).
 
 </Tip>
 
@@ -788,7 +788,7 @@ make_image_grid([init_image, mask_image, image, image_elden_ring], rows=2, cols=
 
 ## Optimize
 
-It can be difficult and slow to run diffusion models if you're resource constrained, but it doesn't have to be with a few optimization tricks. One of the biggest (and easiest) optimizations you can enable is switching to memory-efficient attention. If you're using PyTorch 2.0, [scaled-dot product attention](../optimization/torch2.0#scaled-dot-product-attention) is automatically enabled and you don't need to do anything else. For non-PyTorch 2.0 users, you can install and use [xFormers](../optimization/xformers)'s implementation of memory-efficient attention. Both options reduce memory usage and accelerate inference.
+It can be difficult and slow to run diffusion models if you're resource constrained, but it doesn't have to be with a few optimization tricks. One of the biggest (and easiest) optimizations you can enable is switching to memory-efficient attention. If you're using PyTorch 2.0, [scaled-dot product attention](../optimization/fp16#scaled-dot-product-attention) is automatically enabled and you don't need to do anything else. For non-PyTorch 2.0 users, you can install and use [xFormers](../optimization/xformers)'s implementation of memory-efficient attention. Both options reduce memory usage and accelerate inference.
 
 You can also offload the model to the CPU to save even more memory:
 
@@ -797,10 +797,10 @@ You can also offload the model to the CPU to save even more memory:
 + pipeline.enable_model_cpu_offload()
 ```
 
-To speed-up your inference code even more, use [`torch_compile`](../optimization/torch2.0#torchcompile). You should wrap `torch.compile` around the most intensive component in the pipeline which is typically the UNet:
+To speed-up your inference code even more, use [`torch_compile`](../optimization/fp16#torchcompile). You should wrap `torch.compile` around the most intensive component in the pipeline which is typically the UNet:
 
 ```py
 pipeline.unet = torch.compile(pipeline.unet, mode="reduce-overhead", fullgraph=True)
 ```
 
-Learn more in the [Reduce memory usage](../optimization/memory) and [Torch 2.0](../optimization/torch2.0) guides.
+Learn more in the [Reduce memory usage](../optimization/memory) and [Accelerate inference](../optimization/fp16) guides.

docs/source/en/using-diffusers/marigold_usage.md

@@ -288,7 +288,7 @@ Speeding them up can be achieved by using a more efficient attention processor:
   depth = pipe(image, num_inference_steps=1)
 ```
 
-Finally, as suggested in [Optimizations](../optimization/torch2.0#torch.compile), enabling `torch.compile` can further enhance performance depending on 
+Finally, as suggested in [Optimizations](../optimization/fp16#torchcompile), enabling `torch.compile` can further enhance performance depending on 
 the target hardware.
 However, compilation incurs a significant overhead during the first pipeline invocation, making it beneficial only when 
 the same pipeline instance is called repeatedly, such as within a loop.

docs/source/en/using-diffusers/svd.md

@@ -63,7 +63,7 @@ export_to_video(frames, "generated.mp4", fps=7)
 
 ## torch.compile
 
-You can gain a 20-25% speedup at the expense of slightly increased memory by [compiling](../optimization/torch2.0#torchcompile) the UNet.
+You can gain a 20-25% speedup at the expense of slightly increased memory by [compiling](../optimization/fp16#torchcompile) the UNet.
 
 ```diff
 - pipe.enable_model_cpu_offload()

docs/source/en/using-diffusers/text-img2vid.md

@@ -547,7 +547,7 @@ Video generation requires a lot of memory because you're generating many video f
 + frames = pipeline(image, decode_chunk_size=2, generator=generator, num_frames=25).frames[0]
 ```
 
-If memory is not an issue and you want to optimize for speed, try wrapping the UNet with [`torch.compile`](../optimization/torch2.0#torchcompile).
+If memory is not an issue and you want to optimize for speed, try wrapping the UNet with [`torch.compile`](../optimization/fp16#torchcompile).
 
 ```diff
 - pipeline.enable_model_cpu_offload()