[docs] Add safetensors flag (huggingface#4245)

* add safetensors flag

* apply review

Author: stevhliu

Diff for: docs/source/en/optimization/fp16.md

@@ -51,6 +51,7 @@ from diffusers import DiffusionPipeline
 pipe = DiffusionPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",
     torch_dtype=torch.float16,
+    use_safetensors=True,
 )
 pipe = pipe.to("cuda")
 
@@ -80,6 +81,7 @@ from diffusers import StableDiffusionPipeline
 pipe = StableDiffusionPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",
     torch_dtype=torch.float16,
+    use_safetensors=True,
 )
 pipe = pipe.to("cuda")
 
@@ -106,6 +108,7 @@ from diffusers import StableDiffusionPipeline, UniPCMultistepScheduler
 pipe = StableDiffusionPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",
     torch_dtype=torch.float16,
+    use_safetensors=True,
 )
 pipe.scheduler = UniPCMultistepScheduler.from_config(pipe.scheduler.config)
 pipe = pipe.to("cuda")
@@ -133,6 +136,7 @@ from diffusers import StableDiffusionPipeline
 pipe = StableDiffusionPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",
     torch_dtype=torch.float16,
+    use_safetensors=True,
 )
 
 prompt = "a photo of an astronaut riding a horse on mars"
@@ -157,6 +161,7 @@ from diffusers import StableDiffusionPipeline
 pipe = StableDiffusionPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",
     torch_dtype=torch.float16,
+    use_safetensors=True,
 )
 
 prompt = "a photo of an astronaut riding a horse on mars"
@@ -189,6 +194,7 @@ from diffusers import StableDiffusionPipeline
 pipe = StableDiffusionPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",  
     torch_dtype=torch.float16,
+    use_safetensors=True,
 )
 
 prompt = "a photo of an astronaut riding a horse on mars"
@@ -205,6 +211,7 @@ from diffusers import StableDiffusionPipeline
 pipe = StableDiffusionPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",
     torch_dtype=torch.float16,
+    use_safetensors=True,
 )
 
 prompt = "a photo of an astronaut riding a horse on mars"
@@ -267,6 +274,7 @@ def generate_inputs():
 pipe = StableDiffusionPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",
     torch_dtype=torch.float16,
+    use_safetensors=True,
 ).to("cuda")
 unet = pipe.unet
 unet.eval()
@@ -330,6 +338,7 @@ class UNet2DConditionOutput:
 pipe = StableDiffusionPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",
     torch_dtype=torch.float16,
+    use_safetensors=True,
 ).to("cuda")
 
 # use jitted unet
@@ -389,6 +398,7 @@ import torch
 pipe = DiffusionPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",
     torch_dtype=torch.float16,
+    use_safetensors=True,
 ).to("cuda")
 
 pipe.enable_xformers_memory_efficient_attention()

Diff for: docs/source/en/optimization/torch2.0.md

@@ -39,7 +39,7 @@ pip install --upgrade torch diffusers
     import torch
     from diffusers import DiffusionPipeline
 
-    pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16)
+    pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True)
     pipe = pipe.to("cuda")
 
     prompt = "a photo of an astronaut riding a horse on mars"
@@ -53,7 +53,7 @@ pip install --upgrade torch diffusers
     from diffusers import DiffusionPipeline
     + from diffusers.models.attention_processor import AttnProcessor2_0
 
-    pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
+    pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
     + pipe.unet.set_attn_processor(AttnProcessor2_0())
 
     prompt = "a photo of an astronaut riding a horse on mars"
@@ -69,7 +69,7 @@ pip install --upgrade torch diffusers
     from diffusers import DiffusionPipeline
     from diffusers.models.attention_processor import AttnProcessor
 
-    pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16).to("cuda")
+    pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True).to("cuda")
     pipe.unet.set_default_attn_processor()
 
     prompt = "a photo of an astronaut riding a horse on mars"
@@ -107,7 +107,7 @@ path = "runwayml/stable-diffusion-v1-5"
 
 run_compile = True  # Set True / False
 
-pipe = DiffusionPipeline.from_pretrained(path, torch_dtype=torch.float16)
+pipe = DiffusionPipeline.from_pretrained(path, torch_dtype=torch.float16, use_safetensors=True)
 pipe = pipe.to("cuda")
 pipe.unet.to(memory_format=torch.channels_last)
 
@@ -140,7 +140,7 @@ path = "runwayml/stable-diffusion-v1-5"
 
 run_compile = True  # Set True / False
 
-pipe = StableDiffusionImg2ImgPipeline.from_pretrained(path, torch_dtype=torch.float16)
+pipe = StableDiffusionImg2ImgPipeline.from_pretrained(path, torch_dtype=torch.float16, use_safetensors=True)
 pipe = pipe.to("cuda")
 pipe.unet.to(memory_format=torch.channels_last)
 
@@ -180,7 +180,7 @@ path = "runwayml/stable-diffusion-inpainting"
 
 run_compile = True  # Set True / False
 
-pipe = StableDiffusionInpaintPipeline.from_pretrained(path, torch_dtype=torch.float16)
+pipe = StableDiffusionInpaintPipeline.from_pretrained(path, torch_dtype=torch.float16, use_safetensors=True)
 pipe = pipe.to("cuda")
 pipe.unet.to(memory_format=torch.channels_last)
 
@@ -212,9 +212,9 @@ init_image = init_image.resize((512, 512))
 path = "runwayml/stable-diffusion-v1-5"
 
 run_compile = True  # Set True / False
-controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-canny", torch_dtype=torch.float16)
+controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-canny", torch_dtype=torch.float16, use_safetensors=True)
 pipe = StableDiffusionControlNetPipeline.from_pretrained(
-    path, controlnet=controlnet, torch_dtype=torch.float16
+    path, controlnet=controlnet, torch_dtype=torch.float16, use_safetensors=True
 )
 
 pipe = pipe.to("cuda")
@@ -240,11 +240,11 @@ import torch
 
 run_compile = True  # Set True / False
 
-pipe = DiffusionPipeline.from_pretrained("DeepFloyd/IF-I-M-v1.0", variant="fp16", text_encoder=None, torch_dtype=torch.float16)
+pipe = DiffusionPipeline.from_pretrained("DeepFloyd/IF-I-M-v1.0", variant="fp16", text_encoder=None, torch_dtype=torch.float16, use_safetensors=True)
 pipe.to("cuda")
-pipe_2 = DiffusionPipeline.from_pretrained("DeepFloyd/IF-II-M-v1.0", variant="fp16", text_encoder=None, torch_dtype=torch.float16)
+pipe_2 = DiffusionPipeline.from_pretrained("DeepFloyd/IF-II-M-v1.0", variant="fp16", text_encoder=None, torch_dtype=torch.float16, use_safetensors=True)
 pipe_2.to("cuda")
-pipe_3 = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-x4-upscaler", torch_dtype=torch.float16)
+pipe_3 = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-x4-upscaler", torch_dtype=torch.float16, use_safetensors=True)
 pipe_3.to("cuda")
 
 

Diff for: docs/source/en/quicktour.md

@@ -67,7 +67,7 @@ Load the model with the [`~DiffusionPipeline.from_pretrained`] method:
 ```python
 >>> from diffusers import DiffusionPipeline
 
->>> pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5")
+>>> pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
 ```
 
 The [`DiffusionPipeline`] downloads and caches all modeling, tokenization, and scheduling components. You'll see that the Stable Diffusion pipeline is composed of the [`UNet2DConditionModel`] and [`PNDMScheduler`] among other things:
@@ -130,7 +130,7 @@ You can also use the pipeline locally. The only difference is you need to downlo
 Then load the saved weights into the pipeline:
 
 ```python
->>> pipeline = DiffusionPipeline.from_pretrained("./stable-diffusion-v1-5")
+>>> pipeline = DiffusionPipeline.from_pretrained("./stable-diffusion-v1-5", use_safetensors=True)
 ```
 
 Now you can run the pipeline as you would in the section above.
@@ -142,7 +142,7 @@ Different schedulers come with different denoising speeds and quality trade-offs
 ```py
 >>> from diffusers import EulerDiscreteScheduler
 
->>> pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5")
+>>> pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
 >>> pipeline.scheduler = EulerDiscreteScheduler.from_config(pipeline.scheduler.config)
 ```
 
@@ -160,7 +160,7 @@ Models are initiated with the [`~ModelMixin.from_pretrained`] method which also
 >>> from diffusers import UNet2DModel
 
 >>> repo_id = "google/ddpm-cat-256"
->>> model = UNet2DModel.from_pretrained(repo_id)
+>>> model = UNet2DModel.from_pretrained(repo_id, use_safetensors=True)
 ```
 
 To access the model parameters, call `model.config`:

Diff for: docs/source/en/stable_diffusion.md

@@ -26,7 +26,7 @@ Begin by loading the [`runwayml/stable-diffusion-v1-5`](https://huggingface.co/r
 from diffusers import DiffusionPipeline
 
 model_id = "runwayml/stable-diffusion-v1-5"
-pipeline = DiffusionPipeline.from_pretrained(model_id)
+pipeline = DiffusionPipeline.from_pretrained(model_id, use_safetensors=True)
 ```
 
 The example prompt you'll use is a portrait of an old warrior chief, but feel free to use your own prompt:
@@ -75,7 +75,7 @@ Let's start by loading the model in `float16` and generate an image:
 ```python
 import torch
 
-pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
+pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16, use_safetensors=True)
 pipeline = pipeline.to("cuda")
 generator = torch.Generator("cuda").manual_seed(0)
 image = pipeline(prompt, generator=generator).images[0]

Diff for: docs/source/en/training/adapt_a_model.md

@@ -11,7 +11,7 @@ A [`UNet2DConditionModel`] by default accepts 4 channels in the [input sample](h
 ```py
 from diffusers import StableDiffusionPipeline
 
-pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5")
+pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", use_safetensors=True)
 pipeline.unet.config["in_channels"]
 4
 ```
@@ -21,7 +21,7 @@ Inpainting requires 9 channels in the input sample. You can check this value in
 ```py
 from diffusers import StableDiffusionPipeline
 
-pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-inpainting")
+pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-inpainting", use_safetensors=True)
 pipeline.unet.config["in_channels"]
 9
 ```
@@ -35,7 +35,12 @@ from diffusers import UNet2DConditionModel
 
 model_id = "runwayml/stable-diffusion-v1-5"
 unet = UNet2DConditionModel.from_pretrained(
-    model_id, subfolder="unet", in_channels=9, low_cpu_mem_usage=False, ignore_mismatched_sizes=True
+    model_id,
+    subfolder="unet",
+    in_channels=9,
+    low_cpu_mem_usage=False,
+    ignore_mismatched_sizes=True,
+    use_safetensors=True,
 )
 ```
 

Diff for: docs/source/en/training/controlnet.md

@@ -306,9 +306,9 @@ import torch
 base_model_path = "path to model"
 controlnet_path = "path to controlnet"
 
-controlnet = ControlNetModel.from_pretrained(controlnet_path, torch_dtype=torch.float16)
+controlnet = ControlNetModel.from_pretrained(controlnet_path, torch_dtype=torch.float16, use_safetensors=True)
 pipe = StableDiffusionControlNetPipeline.from_pretrained(
-    base_model_path, controlnet=controlnet, torch_dtype=torch.float16
+    base_model_path, controlnet=controlnet, torch_dtype=torch.float16, use_safetensors=True
 )
 
 # speed up diffusion process with faster scheduler and memory optimization

Diff for: docs/source/en/training/custom_diffusion.md

@@ -222,7 +222,9 @@ Once you have trained a model using the above command, you can run inference usi
 import torch
 from diffusers import DiffusionPipeline
 
-pipe = DiffusionPipeline.from_pretrained("CompVis/stable-diffusion-v1-4", torch_dtype=torch.float16).to("cuda")
+pipe = DiffusionPipeline.from_pretrained(
+    "CompVis/stable-diffusion-v1-4", torch_dtype=torch.float16, use_safetensors=True
+).to("cuda")
 pipe.unet.load_attn_procs("path-to-save-model", weight_name="pytorch_custom_diffusion_weights.bin")
 pipe.load_textual_inversion("path-to-save-model", weight_name="<new1>.bin")
 
@@ -246,7 +248,7 @@ model_id = "sayakpaul/custom-diffusion-cat"
 card = RepoCard.load(model_id)
 base_model_id = card.data.to_dict()["base_model"]
 
-pipe = DiffusionPipeline.from_pretrained(base_model_id, torch_dtype=torch.float16).to("cuda")
+pipe = DiffusionPipeline.from_pretrained(base_model_id, torch_dtype=torch.float16, use_safetensors=True).to("cuda")
 pipe.unet.load_attn_procs(model_id, weight_name="pytorch_custom_diffusion_weights.bin")
 pipe.load_textual_inversion(model_id, weight_name="<new1>.bin")
 
@@ -270,7 +272,7 @@ model_id = "sayakpaul/custom-diffusion-cat-wooden-pot"
 card = RepoCard.load(model_id)
 base_model_id = card.data.to_dict()["base_model"]
 
-pipe = DiffusionPipeline.from_pretrained(base_model_id, torch_dtype=torch.float16).to("cuda")
+pipe = DiffusionPipeline.from_pretrained(base_model_id, torch_dtype=torch.float16, use_safetensors=True).to("cuda")
 pipe.unet.load_attn_procs(model_id, weight_name="pytorch_custom_diffusion_weights.bin")
 pipe.load_textual_inversion(model_id, weight_name="<new1>.bin")
 pipe.load_textual_inversion(model_id, weight_name="<new2>.bin")

Diff for: docs/source/en/training/distributed_inference.md

@@ -16,7 +16,9 @@ Now use the [`~accelerate.PartialState.split_between_processes`] utility as a co
 from accelerate import PartialState
 from diffusers import DiffusionPipeline
 
-pipeline = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16)
+pipeline = DiffusionPipeline.from_pretrained(
+    "runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
+)
 distributed_state = PartialState()
 pipeline.to(distributed_state.device)
 
@@ -50,7 +52,9 @@ import torch.multiprocessing as mp
 
 from diffusers import DiffusionPipeline
 
-sd = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16)
+sd = DiffusionPipeline.from_pretrained(
+    "runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16, use_safetensors=True
+)
 ```
 
 You'll want to create a function to run inference; [`init_process_group`](https://pytorch.org/docs/stable/distributed.html?highlight=init_process_group#torch.distributed.init_process_group) handles creating a distributed environment with the type of backend to use, the `rank` of the current process, and the `world_size` or the number of processes participating. If you're running inference in parallel over 2 GPUs, then the `world_size` is 2.

Diff for: docs/source/en/training/dreambooth.md

@@ -303,7 +303,9 @@ unet = UNet2DConditionModel.from_pretrained("/sddata/dreambooth/daruma-v2-1/chec
 # if you have trained with `--args.train_text_encoder` make sure to also load the text encoder
 text_encoder = CLIPTextModel.from_pretrained("/sddata/dreambooth/daruma-v2-1/checkpoint-100/text_encoder")
 
-pipeline = DiffusionPipeline.from_pretrained(model_id, unet=unet, text_encoder=text_encoder, dtype=torch.float16)
+pipeline = DiffusionPipeline.from_pretrained(
+    model_id, unet=unet, text_encoder=text_encoder, dtype=torch.float16, use_safetensors=True
+)
 pipeline.to("cuda")
 
 # Perform inference, or save, or push to the hub
@@ -318,7 +320,7 @@ from diffusers import DiffusionPipeline
 
 # Load the pipeline with the same arguments (model, revision) that were used for training
 model_id = "CompVis/stable-diffusion-v1-4"
-pipeline = DiffusionPipeline.from_pretrained(model_id)
+pipeline = DiffusionPipeline.from_pretrained(model_id, use_safetensors=True)
 
 accelerator = Accelerator()
 
@@ -333,6 +335,7 @@ pipeline = DiffusionPipeline.from_pretrained(
     model_id,
     unet=accelerator.unwrap_model(unet),
     text_encoder=accelerator.unwrap_model(text_encoder),
+    use_safetensors=True,
 )
 
 # Perform inference, or save, or push to the hub
@@ -488,7 +491,7 @@ from diffusers import DiffusionPipeline
 import torch
 
 model_id = "path_to_saved_model"
-pipe = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to("cuda")
+pipe = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16, use_safetensors=True).to("cuda")
 
 prompt = "A photo of sks dog in a bucket"
 image = pipe(prompt, num_inference_steps=50, guidance_scale=7.5).images[0]
@@ -510,7 +513,7 @@ must also update the pipeline's scheduler config.
 ```py
 from diffusers import DiffusionPipeline
 
-pipe = DiffusionPipeline.from_pretrained("DeepFloyd/IF-I-XL-v1.0")
+pipe = DiffusionPipeline.from_pretrained("DeepFloyd/IF-I-XL-v1.0", use_safetensors=True)
 
 pipe.load_lora_weights("<lora weights path>")
 

Diff for: docs/source/en/training/instructpix2pix.md

@@ -165,7 +165,9 @@ import torch
 from diffusers import StableDiffusionInstructPix2PixPipeline
 
 model_id = "your_model_id"  # <- replace this
-pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to("cuda")
+pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(
+    model_id, torch_dtype=torch.float16, use_safetensors=True
+).to("cuda")
 generator = torch.Generator("cuda").manual_seed(0)
 
 url = "https://huggingface.co/datasets/sayakpaul/sample-datasets/resolve/main/test_pix2pix_4.png"