Update app.py

app.py

@@ -14,7 +14,7 @@ with open('loras.json', 'r') as f:
     loras = json.load(f)
 
 # Initialize the base model
-base_model = "AlekseyCalvin/MythsticColorV2FluxSoonr"
+base_model = "stabilityai/stable-diffusion-3.5-large-turbo"
 pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
 
 MAX_SEED = 2**32-1
@@ -57,7 +57,7 @@ def update_selection(evt: gr.SelectData, width, height):
     )
 
 @spaces.GPU(duration=70)
-def generate_image(prompt, trigger_word, steps, seed, cfg_scale, width, height, lora_scale, progress):
+def infer(prompt, negative_prompt, trigger_word, steps, seed, cfg_scale, width, height, lora_scale, progress):
     pipe.to("cuda")
     generator = torch.Generator(device="cuda").manual_seed(seed)
     
@@ -65,6 +65,7 @@ def generate_image(prompt, trigger_word, steps, seed, cfg_scale, width, height,
         # Generate image
         image = pipe(
             prompt=f"{prompt} {trigger_word}",
+            negative_prompt=negative_prompt,
             num_inference_steps=steps,
             guidance_scale=cfg_scale,
             width=width,
@@ -74,7 +75,7 @@ def generate_image(prompt, trigger_word, steps, seed, cfg_scale, width, height,
         ).images[0]
     return image
 
-def run_lora(prompt, cfg_scale, steps, selected_index, randomize_seed, seed, width, height, lora_scale, progress=gr.Progress(track_tqdm=True)):
+def run_lora(prompt, negative_prompt, cfg_scale, steps, selected_index, randomize_seed, seed, width, height, lora_scale, progress=gr.Progress(track_tqdm=True)):
     if selected_index is None:
         raise gr.Error("You must select a LoRA before proceeding.")
 
@@ -94,7 +95,7 @@ def run_lora(prompt, cfg_scale, steps, selected_index, randomize_seed, seed, wid
         if randomize_seed:
             seed = random.randint(0, MAX_SEED)
     
-    image = generate_image(prompt, trigger_word, steps, seed, cfg_scale, width, height, lora_scale, progress)
+    image = infer(prompt, negative_prompt, trigger_word, steps, seed, cfg_scale, width, height, lora_scale, progress)
     pipe.to("cpu")
     pipe.unload_lora_weights()
     return image, seed  
@@ -110,22 +111,24 @@ css = '''
 '''
 with gr.Blocks(theme=gr.themes.Soft(), css=css) as app:
     title = gr.HTML(
-        """<h1><img src="https://huggingface.co/spaces/multimodalart/flux-lora-the-explorer/resolve/main/flux_lora.png" alt="LoRA"> SOONfactory </h1>""",
+        """<h1><img src="https://huggingface.co/AlekseyCalvin/HSTklimbimOPENfluxLora/resolve/main/acs62iv.png" alt="LoRA">OpenFlux LoRAsoon®</h1>""",
         elem_id="title",
     )
     	    # Info blob stating what the app is running
     info_blob = gr.HTML(
-        """<div id="info_blob"> Img. Manufactory Running On: Our 'Mythstic Color Soonr V.2' 4-step(+/-) Flux Model (at AlekseyCalvin/AlekseyCalvin/MythsticColorV2FluxSoonr and AlekseyCalvin/Mythstic_Color_Soonr_Flux_V2 (for Safetensors version)). Now testing HST-triggerable historic photo-trained LoRAs (#s:2-8,11,12,14,16)for training-eval merging. </div>"""
+        """<div id="info_blob"> SOON®'s curated LoRa Gallery & Art Manufactory Space.|Runs on Stable Diffusion 3.5 Turbo. Now testing HST-triggerable historic photo-trained LoRAs. </div>"""
     )
 
         # Info blob stating what the app is running
     info_blob = gr.HTML(
-        """<div id="info_blob">Prephrase prompts w/: 1: RCA style || 2-thru-12: HST style analog film photo; HST autochrome photograph || 13: HST style in Peterhof || 14: LEN Vladimir Lenin || 15: SOTS style || 16: crisp photo || 17: TOK hybrid || 18: 2004 photo || 19: TOK portra || 20: flmft Kodachrome || 21: HST Austin Osman Spare style || 22: polaroid photo || 23: pficonics || 24: wh3r3sw4ld0 || 25: retrofuturism || 26: vintage cover || </div>"""
+        """<div id="info_blob">Prephrase prompts w/: "HST style autochrome photo" </div>"""
     )
     selected_index = gr.State(None)
     with gr.Row():
-        with gr.Column(scale=3):
+        with gr.Column(scale=2):
             prompt = gr.Textbox(label="Prompt", lines=1, placeholder="Select LoRa/Style & type prompt!")
+        with gr.Column(scale=2):
+            negative_prompt = gr.Textbox(label="Negative Prompt", lines=1, placeholder="What to exclude!")
         with gr.Column(scale=1, elem_id="gen_column"):
             generate_button = gr.Button("Generate", variant="primary", elem_id="gen_btn")
     with gr.Row():
@@ -146,17 +149,17 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css) as app:
         with gr.Accordion("Advanced Settings", open=True):
             with gr.Column():
                 with gr.Row():
-                    cfg_scale = gr.Slider(label="CFG Scale", minimum=0, maximum=20, step=.5, value=0)
+                    cfg_scale = gr.Slider(label="CFG Scale", minimum=0, maximum=20, step=.5, value=4)
                     steps = gr.Slider(label="Steps", minimum=1, maximum=50, step=1, value=4)
                 
                 with gr.Row():
-                    width = gr.Slider(label="Width", minimum=256, maximum=1536, step=64, value=768)
-                    height = gr.Slider(label="Height", minimum=256, maximum=1536, step=64, value=768)
+                    width = gr.Slider(label="Width", minimum=256, maximum=1536, step=64, value=1024)
+                    height = gr.Slider(label="Height", minimum=256, maximum=1536, step=64, value=1024)
                 
                 with gr.Row():
                     randomize_seed = gr.Checkbox(True, label="Randomize seed")
                     seed = gr.Slider(label="Seed", minimum=0, maximum=MAX_SEED, step=1, value=0, randomize=True)
-                    lora_scale = gr.Slider(label="LoRA Scale", minimum=0, maximum=3.0, step=0.01, value=0.5)
+                    lora_scale = gr.Slider(label="LoRA Scale", minimum=0, maximum=3.0, step=0.01, value=1.0)
 
     gallery.select(
         update_selection,
@@ -167,7 +170,7 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css) as app:
     gr.on(
         triggers=[generate_button.click, prompt.submit],
         fn=run_lora,
-        inputs=[prompt, cfg_scale, steps, selected_index, randomize_seed, seed, width, height, lora_scale],
+        inputs=[prompt, negative_prompt, cfg_scale, steps, selected_index, randomize_seed, seed, width, height, lora_scale],
         outputs=[result, seed]
     )