Update app.py

app.py

@@ -41,7 +41,7 @@ def generate_response(prompt, max_tokens=150, temperature=0.7):
     Uses LLAMA-CPP to generate a response for the given prompt.
     Note: Removed echo=True to avoid repeating the prompt.
     """
-    output = llm(prompt, max_tokens=max_tokens, temperature=temperature)  # echo removed
+    output = llm(prompt, max_tokens=max_tokens, temperature=temperature, stop=["User:"])  # echo removed
     response = output["choices"][0]["text"]
     return response.strip()