Update app.py

app.py

@@ -1,154 +1,99 @@
 import gradio as gr
-import numpy as np
-import random
-
-# import spaces #[uncomment to use ZeroGPU]
-from diffusers import DiffusionPipeline
-import torch
-
-device = "cuda" if torch.cuda.is_available() else "cpu"
-model_repo_id = "stabilityai/sdxl-turbo"  # Replace to the model you would like to use
-
-if torch.cuda.is_available():
-    torch_dtype = torch.float16
-else:
-    torch_dtype = torch.float32
-
-pipe = DiffusionPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype)
-pipe = pipe.to(device)
-
-MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 1024
-
-
-# @spaces.GPU #[uncomment to use ZeroGPU]
-def infer(
-    prompt,
-    negative_prompt,
-    seed,
-    randomize_seed,
-    width,
-    height,
-    guidance_scale,
-    num_inference_steps,
-    progress=gr.Progress(track_tqdm=True),
-):
-    if randomize_seed:
-        seed = random.randint(0, MAX_SEED)
-
-    generator = torch.Generator().manual_seed(seed)
-
-    image = pipe(
-        prompt=prompt,
-        negative_prompt=negative_prompt,
-        guidance_scale=guidance_scale,
-        num_inference_steps=num_inference_steps,
-        width=width,
-        height=height,
-        generator=generator,
-    ).images[0]
-
-    return image, seed
-
-
-examples = [
-    "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
-    "An astronaut riding a green horse",
-    "A delicious ceviche cheesecake slice",
-]
-
-css = """
-#col-container {
-    margin: 0 auto;
-    max-width: 640px;
-}
-"""
-
-with gr.Blocks(css=css) as demo:
-    with gr.Column(elem_id="col-container"):
-        gr.Markdown(" # Text-to-Image Gradio Template")
-
-        with gr.Row():
-            prompt = gr.Text(
-                label="Prompt",
-                show_label=False,
-                max_lines=1,
-                placeholder="Enter your prompt",
-                container=False,
-            )
-
-            run_button = gr.Button("Run", scale=0, variant="primary")
-
-        result = gr.Image(label="Result", show_label=False)
-
-        with gr.Accordion("Advanced Settings", open=False):
-            negative_prompt = gr.Text(
-                label="Negative prompt",
-                max_lines=1,
-                placeholder="Enter a negative prompt",
-                visible=False,
-            )
-
-            seed = gr.Slider(
-                label="Seed",
-                minimum=0,
-                maximum=MAX_SEED,
-                step=1,
-                value=0,
-            )
-
-            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
-
-            with gr.Row():
-                width = gr.Slider(
-                    label="Width",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=1024,  # Replace with defaults that work for your model
-                )
-
-                height = gr.Slider(
-                    label="Height",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=1024,  # Replace with defaults that work for your model
-                )
-
-            with gr.Row():
-                guidance_scale = gr.Slider(
-                    label="Guidance scale",
-                    minimum=0.0,
-                    maximum=10.0,
-                    step=0.1,
-                    value=0.0,  # Replace with defaults that work for your model
-                )
-
-                num_inference_steps = gr.Slider(
-                    label="Number of inference steps",
-                    minimum=1,
-                    maximum=50,
-                    step=1,
-                    value=2,  # Replace with defaults that work for your model
-                )
-
-        gr.Examples(examples=examples, inputs=[prompt])
-    gr.on(
-        triggers=[run_button.click, prompt.submit],
-        fn=infer,
-        inputs=[
-            prompt,
-            negative_prompt,
-            seed,
-            randomize_seed,
-            width,
-            height,
-            guidance_scale,
-            num_inference_steps,
-        ],
-        outputs=[result, seed],
+import base64
+import io
+import os
+from PIL import Image
+from openai import OpenAI
+#from google.colab import userdata
+
+# 1. Setup Client
+try:
+    api_key = os.environ.get('HYPERBOLIC_API_KEY')
+    client = OpenAI(
+        api_key=api_key,
+        base_url="https://api.hyperbolic.xyz/v1"
     )
-
-if __name__ == "__main__":
-    demo.launch()
+except Exception as e:
+    print(f"Secret-Fehler: {e}")
+
+# 2. Optimierte Bild-Funktion (DAS IST NEU)
+def encode_image_optimized(image_path, max_size=1024):
+    """
+    Öffnet das Bild, verkleinert es, wenn es zu riesig ist,
+    und gibt es als Base64 zurück. Verhindert Browser-Abstürze.
+    """
+    try:
+        with Image.open(image_path) as img:
+            # Konvertiere zu RGB (falls PNG mit Transparenz, was oft Fehler macht)
+            if img.mode in ("RGBA", "P"):
+                img = img.convert("RGB")
+
+            # Skalieren, wenn größer als max_size (z.B. > 1024px)
+            if max(img.size) > max_size:
+                img.thumbnail((max_size, max_size))
+
+            # In Buffer speichern statt auf Festplatte
+            buffered = io.BytesIO()
+            img.save(buffered, format="JPEG", quality=85) # Quality 85 spart massiv Platz
+            return base64.b64encode(buffered.getvalue()).decode('utf-8')
+    except Exception as e:
+        print(f"Fehler beim Bild-Verarbeiten: {e}")
+        return None
+
+# 3. Chat Logik
+def chat_response(message, history):
+    messages_payload = []
+
+    # System Prompt (optional, macht das Modell stabiler)
+    messages_payload.append({"role": "system", "content": "Du bist ein hilfreicher Assistent."})
+
+    # A. Nur Text-Historie übernehmen (Bilder aus der Vergangenheit weglassen)
+    # Warum? Alte Bilder erneut hochzuladen frisst Bandbreite und crasht oft.
+    # Wir senden nur das *aktuelle* Bild + den Text-Kontext der Unterhaltung.
+    for msg in history:
+        if isinstance(msg["content"], str):
+             messages_payload.append({"role": msg["role"], "content": msg["content"]})
+
+    # B. Aktueller Input
+    current_content = []
+
+    # Text
+    if message["text"]:
+        current_content.append({"type": "text", "text": message["text"]})
+
+    # Bild (Optimiert!)
+    if message["files"]:
+        for file_path in message["files"]:
+            b64_img = encode_image_optimized(file_path) # Hier nutzen wir die neue Funktion
+            if b64_img:
+                current_content.append({
+                    "type": "image_url",
+                    "image_url": {"url": f"data:image/jpeg;base64,{b64_img}"}
+                })
+
+    messages_payload.append({"role": "user", "content": current_content})
+
+    # C. API Call
+    try:
+        response = client.chat.completions.create(
+            model="nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-BF16",
+            messages=messages_payload,
+            max_tokens=1024,
+            temperature=0.7
+        )
+        return response.choices[0].message.content
+
+    except Exception as e:
+        return f"API Fehler: {str(e)}"
+
+# 4. UI Starten
+demo = gr.ChatInterface(
+    fn=chat_response,
+    #type="messages",
+    multimodal=True,
+    title="Hyperbolic Vision (Optimiert)",
+    description="Läuft stabil auch mit großen Bildern (Auto-Resize).",
+    textbox=gr.MultimodalTextbox(placeholder="Bild reinziehen...", file_count="multiple"),
+)
+
+demo.launch(share=True, debug=True)