[Sana Sprint] add image-to-image pipeline (#11602)

* sana sprint img2img

* fix import

* fix name

* fix image encoding

* fix image encoding

* fix image encoding

* fix image encoding

* fix image encoding

* fix image encoding

* try w/o strength

* try scaling differently

* try with strength

* revert unnecessary changes to scheduler

* revert unnecessary changes to scheduler

* Apply style fixes

* remove comment

* add copy statements

* add copy statements

* add to doc

* add to doc

* add to doc

* add to doc

* Apply style fixes

* empty commit

* fix copies

* fix copies

* fix copies

* fix copies

* fix copies

* docs

* make fix-copies.

* fix doc building error.

* initial commit - add img2img test

* initial commit - add img2img test

* fix import

* fix imports

* Apply style fixes

* empty commit

* remove

* empty commit

* test vocab size

* fix

* fix prompt missing from last commits

* small changes

* fix image processing when input is tensor

* fix order

* Apply style fixes

* empty commit

* fix shape

* remove comment

* image processing

* remove comment

* skip vae tiling test for now

* Apply style fixes

* empty commit

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: sayakpaul <[email protected]>

Author: 3 people

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

@@ -88,12 +88,46 @@ image.save("sana.png")
 
 Users can tweak the `max_timesteps` value for experimenting with the visual quality of the generated outputs. The default `max_timesteps` value was obtained with an inference-time search process. For more details about it, check out the paper.
 
+## Image to Image 
+
+The [`SanaSprintImg2ImgPipeline`] is a pipeline for image-to-image generation. It takes an input image and a prompt, and generates a new image based on the input image and the prompt.
+
+```py
+import torch
+from diffusers import SanaSprintImg2ImgPipeline
+from diffusers.utils.loading_utils import load_image
+
+image = load_image(
+    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/penguin.png"
+)
+
+pipe = SanaSprintImg2ImgPipeline.from_pretrained(
+    "Efficient-Large-Model/Sana_Sprint_1.6B_1024px_diffusers", 
+    torch_dtype=torch.bfloat16)
+pipe.to("cuda")
+
+image = pipe(
+    prompt="a cute pink bear", 
+    image=image, 
+    strength=0.5, 
+    height=832, 
+    width=480
+).images[0]
+image[0].save("output.png")
+```
+
 ## SanaSprintPipeline
 
 [[autodoc]] SanaSprintPipeline
   - all
   - __call__
 
+## SanaSprintImg2ImgPipeline
+
+[[autodoc]] SanaSprintImg2ImgPipeline
+  - all
+  - __call__
+
 
 ## SanaPipelineOutput
 

src/diffusers/__init__.py

@@ -441,6 +441,7 @@
             "SanaControlNetPipeline",
             "SanaPAGPipeline",
             "SanaPipeline",
+            "SanaSprintImg2ImgPipeline",
             "SanaSprintPipeline",
             "SemanticStableDiffusionPipeline",
             "ShapEImg2ImgPipeline",
@@ -1025,6 +1026,7 @@
             SanaControlNetPipeline,
             SanaPAGPipeline,
             SanaPipeline,
+            SanaSprintImg2ImgPipeline,
             SanaSprintPipeline,
             SemanticStableDiffusionPipeline,
             ShapEImg2ImgPipeline,

src/diffusers/pipelines/__init__.py

@@ -290,7 +290,12 @@
     _import_structure["paint_by_example"] = ["PaintByExamplePipeline"]
     _import_structure["pia"] = ["PIAPipeline"]
     _import_structure["pixart_alpha"] = ["PixArtAlphaPipeline", "PixArtSigmaPipeline"]
-    _import_structure["sana"] = ["SanaPipeline", "SanaSprintPipeline", "SanaControlNetPipeline"]
+    _import_structure["sana"] = [
+        "SanaPipeline",
+        "SanaSprintPipeline",
+        "SanaControlNetPipeline",
+        "SanaSprintImg2ImgPipeline",
+    ]
     _import_structure["semantic_stable_diffusion"] = ["SemanticStableDiffusionPipeline"]
     _import_structure["shap_e"] = ["ShapEImg2ImgPipeline", "ShapEPipeline"]
     _import_structure["stable_audio"] = [
@@ -675,7 +680,7 @@
         from .paint_by_example import PaintByExamplePipeline
         from .pia import PIAPipeline
         from .pixart_alpha import PixArtAlphaPipeline, PixArtSigmaPipeline
-        from .sana import SanaControlNetPipeline, SanaPipeline, SanaSprintPipeline
+        from .sana import SanaControlNetPipeline, SanaPipeline, SanaSprintImg2ImgPipeline, SanaSprintPipeline
         from .semantic_stable_diffusion import SemanticStableDiffusionPipeline
         from .shap_e import ShapEImg2ImgPipeline, ShapEPipeline
         from .stable_audio import StableAudioPipeline, StableAudioProjectionModel

src/diffusers/pipelines/sana/__init__.py

@@ -25,6 +25,7 @@
     _import_structure["pipeline_sana"] = ["SanaPipeline"]
     _import_structure["pipeline_sana_controlnet"] = ["SanaControlNetPipeline"]
     _import_structure["pipeline_sana_sprint"] = ["SanaSprintPipeline"]
+    _import_structure["pipeline_sana_sprint_img2img"] = ["SanaSprintImg2ImgPipeline"]
 
 if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
     try:
@@ -37,6 +38,7 @@
         from .pipeline_sana import SanaPipeline
         from .pipeline_sana_controlnet import SanaControlNetPipeline
         from .pipeline_sana_sprint import SanaSprintPipeline
+        from .pipeline_sana_sprint_img2img import SanaSprintImg2ImgPipeline
 else:
     import sys