Update app.py

app.py

@@ -6,7 +6,7 @@ from huggingface_hub import login
 # Fetch token from environment (automatically loaded from secrets)
 hf_token = os.getenv("gemma3")
 login(hf_token)
-# Initialize the client with your model
+
 client = InferenceClient("hackergeek98/gemma-finetuned")
 
 def respond(
@@ -17,33 +17,27 @@ def respond(
     temperature,
     top_p,
 ):
-    # Preparing the messages list
-    messages = [{"role": "system", "content": system_message}]
-
-    # Adding conversation history
+    prompt = f"{system_message}\n"
+    
+    # Add conversation history if needed
     for val in history:
         if val[0]:
-            messages.append({"role": "user", "content": val[0]})
+            prompt += f"User: {val[0]}\n"
         if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+            prompt += f"Assistant: {val[1]}\n"
 
-    # Adding the new user message
-    messages.append({"role": "user", "content": message})
+    prompt += f"User: {message}\nAssistant:"
 
-    # Prepare the prompt for generation
-    prompt = " ".join([msg["content"] for msg in messages])
-
-    # Call the Inference API for text generation (or chat completion if supported)
-    response = client.completion(
-        model="hackergeek98/gemma-finetuned",  # Specify the model
-        prompt=prompt,
+    # Request generation from Hugging Face Inference API
+    response = client.text_generation(
+        model="hackergeek98/gemma-finetuned",
+        inputs=prompt,
         max_tokens=max_tokens,
         temperature=temperature,
         top_p=top_p,
     )
 
-    # The response will contain the generated text
-    return response["choices"][0]["text"]
+    return response['generated_text']
 
 # Gradio interface setup
 demo = gr.ChatInterface(
@@ -59,3 +53,4 @@ demo = gr.ChatInterface(
 # Run the app
 if __name__ == "__main__":
     demo.launch()
+