Update app.py

app.py

@@ -1,58 +1,65 @@
-# app.py - LOAD ON DEMAND
+# app.py - SIMPLEST WORKING VERSION
 import gradio as gr
 import subprocess
-import tempfile
-import os
+import time
+
+print("🚀 Starting Foundation-Sec-8B Space...")
 
 def generate(prompt):
-    """Load model on-demand using transformers CLI"""
-    # Create a temporary script
-    script = f"""
+    """Generate text using transformers CLI (avoids loading issues)"""
+    python_code = f"""
 import torch
-from transformers import AutoModelForCausalLM, AutoTokenizer
+from transformers import pipeline
 
-model = AutoModelForCausalLM.from_pretrained(
-    "fdtn-ai/Foundation-Sec-8B",
+print("Loading model...")
+model = pipeline(
+    "text-generation",
+    model="fdtn-ai/Foundation-Sec-8B",
+    device=0,
     torch_dtype=torch.float16,
-    device_map="auto",
     trust_remote_code=True
 )
-tokenizer = AutoTokenizer.from_pretrained("fdtn-ai/Foundation-Sec-8B")
 
-inputs = tokenizer('{prompt}', return_tensors="pt").to(model.device)
-outputs = model.generate(**inputs, max_new_tokens=200)
-print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+result = model("{prompt}", max_new_tokens=200, temperature=0.7)
+print(result[0]['generated_text'])
 """
     
-    # Write to temp file
-    with tempfile.NamedTemporaryFile(mode='w', suffix='.py', delete=False) as f:
-        f.write(script)
-        script_path = f.name
+    # Save to temp file
+    with open("/tmp/generate.py", "w") as f:
+        f.write(python_code)
     
     try:
-        # Run script
+        # Run in subprocess
         result = subprocess.run(
-            ['python', script_path],
+            ["python", "/tmp/generate.py"],
             capture_output=True,
             text=True,
-            timeout=120
+            timeout=180
         )
         
-        # Cleanup
-        os.unlink(script_path)
-        
         if result.returncode == 0:
-            return result.stdout.strip()
+            return result.stdout.strip().split('\n')[-1]  # Get last line
         else:
             return f"Error: {result.stderr}"
             
     except subprocess.TimeoutExpired:
-        return "Timeout - Model loading took too long"
+        return "Timeout - Model taking too long to load"
+    except Exception as e:
+        return f"Exception: {str(e)}"
+
+# Create interface
+iface = gr.Interface(
+    fn=generate,
+    inputs=gr.Textbox(label="Ask about cybersecurity:", lines=3),
+    outputs=gr.Textbox(label="Response", lines=10),
+    title="🔒 Foundation-Sec-8B",
+    description="Cybersecurity AI Assistant",
+    examples=[
+        ["What is a firewall?"],
+        ["Explain encryption:"],
+        ["How to create strong passwords?"]
+    ]
+)
 
-# Launch interface
-gr.Interface(
-    generate,
-    gr.Textbox(label="Ask about cybersecurity:"),
-    gr.Textbox(label="Response", lines=10),
-    title="Foundation-Sec-8B (On-demand Loading)"
-).launch(server_name="0.0.0.0")
+if __name__ == "__main__":
+    iface.launch(server_name="0.0.0.0")