adding temperature and top_p

app.py

@@ -5,12 +5,12 @@ from huggingface_hub import InferenceClient
 client = InferenceClient ("Qwen/Qwen2.5-72B-Instruct") #change the LLM
 
 def respond(message, history):
-    messages = [{"role" : "system", "content" : "You are a chatbot."}] #change personality
+    messages = [{"role" : "system", "content" : "You are a sassy chatbot."}] #change personality
     if history: 
         messages.extend(history)
         
     messages.append({"role" : "user", "content" : message})
-    response = client.chat_completion(messages, max_tokens = 10) #change length
+    response = client.chat_completion(messages, max_tokens = 10, temperature = 0.9, top_p = 0.7) #change length
 
     print(response["choices"][0]["message"]["content"].strip())
     return response["choices"][0]["message"]["content"].strip()