Update app.py

response = llm.chat.completions.create(
model=LLM_MODEL,
messages=[{"role": "user", "content": prompt}],
max_tokens=2048,
temperature=0.25,
top_p=0.9,
frequency_penalty=0.1, # reduces repetition
presence_penalty=0.0 # keeps answers grounded
)

app.py

@@ -109,9 +109,11 @@ def generate_answer(question, context):
     response = llm.chat.completions.create(
         model=LLM_MODEL,
         messages=[{"role": "user", "content": prompt}],
-        max_tokens=1500,
-        temperature=0.4,
-        top_p=0.95,
+        max_tokens=2048,
+        temperature=0.25,
+        top_p=0.9,
+        frequency_penalty=0.1,  # reduces repetition
+        presence_penalty=0.0    # keeps answers grounded
     )
 
     return response.choices[0].message["content"].strip()