Create app.py

app.py

@@ -0,0 +1,96 @@
+import subprocess
+import sys
+import os
+
+# Install llama-cpp-python dependency
+def install_llama_cpp():
+    try:
+        print("Installing llama-cpp-python...")
+        subprocess.check_call([
+            sys.executable, '-m', 'pip', 'install', 
+            'llama-cpp-python', '--verbose'
+        ])
+        print("Successfully installed llama-cpp-python")
+        return True
+    except Exception as e:
+        print(f"Error installing llama-cpp-python: {e}")
+        return False
+
+# Install dependency first
+install_success = install_llama_cpp()
+
+if install_success:
+    try:
+        from llama_cpp import Llama
+        import gradio as gr
+        from huggingface_hub import hf_hub_download
+        
+        print("Downloading model from your repository...")
+        
+        # Download the model file from your repo with error handling
+        try:
+            model_path = hf_hub_download(
+                repo_id="Oromor/deepseek-coder-6.7b-instruct.Q4_0.gguf", 
+                filename="deepseek-coder-6.7b-instruct.Q4_0.gguf"
+            )
+            print(f"Model downloaded to: {model_path}")
+        except Exception as e:
+            print(f"Error downloading model: {e}")
+            raise
+        
+        # Load the model with error handling
+        try:
+            print("Loading Llama model...")
+            model = Llama(model_path=model_path, n_ctx=4096, verbose=False)
+            print("Model loaded successfully!")
+        except Exception as e:
+            print(f"Error loading model: {e}")
+            raise
+
+        def predict(input_text):
+            if not input_text.strip():
+                return "Please enter some text."
+            
+            try:
+                print(f"Generating response for: {input_text[:50]}...")
+                output = model(input_text, max_tokens=1500, temperature=0.7, top_p=0.9)
+                
+                if 'choices' in output and len(output['choices']) > 0:
+                    response = output['choices'][0]['text']
+                    return response.strip()
+                else:
+                    return f"Unexpected output format: {str(output)}"
+                    
+            except Exception as e:
+                return f"Error generating response: {str(e)}"
+
+        # Create Gradio interface
+        demo = gr.Interface(
+            fn=predict, 
+            inputs=gr.Textbox(label="Your message", placeholder="Ask me anything about coding..."),
+            outputs=gr.Textbox(label="Deepseek Coder Response"),
+            title="🤖 Deepseek Coder Chat",
+            description="Chat with Deepseek Coder 6.7B model"
+        )
+        
+        print("Starting Gradio interface...")
+        demo.launch()
+        
+    except ImportError as e:
+        import gradio as gr
+        
+        def error_predict(input_text):
+            return f"Failed to import required libraries: {str(e)}"
+        
+        demo = gr.Interface(fn=error_predict, inputs="text", outputs="text", title="Error")
+        demo.launch()
+        
+else:
+    # Fallback if installation fails
+    import gradio as gr
+    
+    def error_predict(input_text):
+        return "Failed to install llama-cpp-python. Please try again or contact support."
+    
+    demo = gr.Interface(fn=error_predict, inputs="text", outputs="text", title="Installation Error")
+    demo.launch()