up

run_local_img2img_xl.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python3
-from diffusers import DiffusionPipeline, EulerDiscreteScheduler, StableDiffusionPipeline, KDPM2DiscreteScheduler, StableDiffusionImg2ImgPipeline, HeunDiscreteScheduler, KDPM2AncestralDiscreteScheduler, DDIMScheduler
+from diffusers import DiffusionPipeline, EulerDiscreteScheduler, StableDiffusionPipeline, KDPM2DiscreteScheduler, StableDiffusionImg2ImgPipeline, HeunDiscreteScheduler, KDPM2AncestralDiscreteScheduler, DDIMScheduler, StableDiffusionXLImg2ImgPipeline
 import time
 import os
 from huggingface_hub import HfApi
@@ -15,23 +15,28 @@ path = sys.argv[1]
 
 api = HfApi()
 start_time = time.time()
-pipe = DiffusionPipeline.from_pretrained(path, torch_dtype=torch.float16)
+pipe = StableDiffusionXLImg2ImgPipeline.from_pretrained(path, torch_dtype=torch.float16)
 pipe.scheduler = EulerDiscreteScheduler.from_config(pipe.scheduler.config)
 # pipe.scheduler = DDIMScheduler.from_config(pipe.scheduler.config)
-# pipe = StableDiffusionImg2ImgPipeline.from_pretrained(path, torch_dtype=torch.float16, safety_checker=None
+# pipe = StableDiffusionImg2ImgXLPipeline.from_pretrained(path, torch_dtype=torch.float16, safety_checker=None
 
 # compel = Compel(tokenizer=pipe.tokenizer, text_encoder=pipe.text_encoder)
 
 
 pipe = pipe.to("cuda")
 
-prompt = "Elon Musk riding a green horse on Mars"
+prompt = "A red castle on a beautiful landscape with a nice sunset"
+
 
 # pipe.unet.to(memory_format=torch.channels_last)
-pipe.unet = torch.compile(pipe.unet, mode="reduce-overhead", fullgraph=True)
-pipe(prompt=prompt, num_inference_steps=2).images[0]
+# pipe.unet = torch.compile(pipe.unet, mode="reduce-overhead", fullgraph=True)
+# pipe(prompt=prompt, num_inference_steps=2).images[0]
+url = "https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg"
+
+response = requests.get(url)
+init_image = Image.open(BytesIO(response.content)).convert("RGB").resize((1024, 1024))
 
-image = pipe(prompt=prompt).images[0]
+image = pipe(prompt=prompt, image=init_image, strength=0.9).images[0]
 
 file_name = f"aaa"
 path = os.path.join(Path.home(), "images", f"{file_name}.png")

run_local_xl.py

@@ -28,10 +28,16 @@ pipe = pipe.to("cuda")
 prompt = "Elon Musk riding a green horse on Mars"
 
 # pipe.unet.to(memory_format=torch.channels_last)
-pipe.unet = torch.compile(pipe.unet, mode="reduce-overhead", fullgraph=True)
-pipe(prompt=prompt, num_inference_steps=2).images[0]
+# pipe.unet = torch.compile(pipe.unet, mode="reduce-overhead", fullgraph=True)
+# pipe(prompt=prompt, num_inference_steps=2).images[0]
 
-image = pipe(prompt=prompt).images[0]
+image = pipe(prompt=prompt, num_images_per_prompt=1, num_inference_steps=40, output_type="latent").images
+pipe.to("cpu")
+
+pipe = DiffusionPipeline.from_pretrained("/home/patrick/diffusers-sd-xl/stable-diffusion-xl-refiner-0.9", torch_dtype=torch.float16)
+pipe.to("cuda")
+
+image = pipe(prompt=prompt, image=image, strength=0.5).images[0]
 
 file_name = f"aaa"
 path = os.path.join(Path.home(), "images", f"{file_name}.png")