Update app.py

app.py

@@ -19,19 +19,10 @@ print(f"[SYSTEM] | Using {DEVICE} type compute device.")
 MAX_SEED = 9007199254740991
 DEFAULT_INPUT = ""
 DEFAULT_NEGATIVE_INPUT = "EasyNegative, deformed, distorted, disfigured, disconnected, disgusting, mutation, mutated, blur, blurry, scribble, abstract, watermark, ugly, amputation, limb, limbs, leg, legs, foot, feet, toe, toes, arm, arms, hand, hands, finger, fingers, head, heads, exposed, porn, nude, nudity, naked, nsfw"
+DEFAULT_MODEL = "None"
 DEFAULT_HEIGHT = 1024
 DEFAULT_WIDTH = 1024
 
-REPO = "hsalf-lxds/ytinummoc-ds"[::-1]
-
-vae = AutoencoderKL.from_pretrained("xif-61pf-eav-lxds/nilloybedam"[::-1], torch_dtype=torch.float16)
-controlnet = ControlNetModel.from_pretrained("k031-sdnah-dedocne-tenlortnoc/naPikaM"[::-1], torch_dtype=torch.float16)
-
-model = StableDiffusionXLPipeline.from_pretrained(REPO, vae=vae, controlnet=controlnet, torch_dtype=torch.float16, use_safetensors=True, add_watermarker=False)
-model.load_lora_weights("2v-lx-3-ellad/urofotsirhe"[::-1], adapter_name="base")
-model.set_adapters(["base"], adapter_weights=[0.7])
-model.to(DEVICE)
-
 css = '''
 .gradio-container{max-width: 560px !important}
 h1{text-align:center}
@@ -54,13 +45,30 @@ def get_seed(seed):
         return random.randint(0, MAX_SEED)
 
 @spaces.GPU(duration=30)
-def generate(input=DEFAULT_INPUT, negative_input=DEFAULT_NEGATIVE_INPUT, height=DEFAULT_HEIGHT, width=DEFAULT_WIDTH, steps=1, guidance=0, number=1, seed=None):
-    
+def generate(input=DEFAULT_INPUT, negative_input=DEFAULT_NEGATIVE_INPUT, model=DEFAULT_MODEL, height=DEFAULT_HEIGHT, width=DEFAULT_WIDTH, steps=1, guidance=0, number=1, seed=None):
+
+    repo = None
     seed = get_seed(seed)
 
-    print(input, negative_input, height, width, steps, guidance, number, seed)
+    print(input, negative_input, model, height, width, steps, guidance, number, seed)
 
+    if model == "Anime":   
+        vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16)
+        controlnet = ControlNetModel.from_pretrained("MakiPan/controlnet-encoded-hands-130k", torch_dtype=torch.float16)
+        repo = StableDiffusionXLPipeline.from_pretrained(cagliostrolab/animagine-xl-3.1, vae=vae, controlnet=controlnet, torch_dtype=torch.float16, use_safetensors=True, add_watermarker=False)
+        steps = steps or 16
+        guidance = guidance or 7
+    else:
+        vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16)
+        controlnet = ControlNetModel.from_pretrained("MakiPan/controlnet-encoded-hands-130k", torch_dtype=torch.float16)
+        repo = StableDiffusionXLPipeline.from_pretrained("sd-community/sdxl-flash", vae=vae, controlnet=controlnet, torch_dtype=torch.float16, use_safetensors=True, add_watermarker=False)
+        repo.load_lora_weights("ehristoforu/dalle-3-xl-v2", adapter_name="base")
+        repo.set_adapters(["base"], adapter_weights=[0.7])
+        steps = steps or 16
+        guidance = guidance or 3
+        
     model.to(DEVICE)
+    
     parameters  = {
         "prompt": input,
         "negative_prompt": negative_input,
@@ -93,6 +101,7 @@ with gr.Blocks(css=css) as main:
     with gr.Column():
         input = gr.Textbox(lines=1, value=DEFAULT_INPUT, label="Input")
         negative_input = gr.Textbox(lines=1, value=DEFAULT_NEGATIVE_INPUT, label="Input Negative")
+        model = gr.Dropdown(label="Models", choices=["Default", "Anime"], value=["Default"])
         height = gr.Slider(minimum=1, maximum=2160, step=1, value=DEFAULT_HEIGHT, label="Height")
         width = gr.Slider(minimum=1, maximum=2160, step=1, value=DEFAULT_WIDTH, label="Width")
         steps = gr.Slider(minimum=0, maximum=100, step=1, value=16, label="Steps")
@@ -105,7 +114,7 @@ with gr.Blocks(css=css) as main:
     with gr.Column():
         images = gr.Gallery(columns=1, label="Image")
             
-    submit.click(generate, inputs=[input, negative_input, height, width, steps, guidance, number, seed], outputs=[images], queue=False)
+    submit.click(generate, inputs=[input, negative_input, model, height, width, steps, guidance, number, seed], outputs=[images], queue=False)
     maintain.click(cloud, inputs=[], outputs=[], queue=False)
 
 main.launch(show_api=True)