modify output length

app.py

@@ -15,15 +15,15 @@ client = OpenAI(
 def chat(user_input, max_new_tokens=20480):
     # chat history
     messages_template = [
-        {"role": "system", "content": "You are Ling, an assistant created by inclusionAI"},
+  #      {"role": "system", "content": "You are Ling, an assistant created by inclusionAI"},
         {"role": "user", "content": user_input}
     ]
 
     response = client.chat.completions.create(
         model="Ling-lite-1.5-250604",
         messages=messages_template,
-        max_tokens=max_new_tokens,
-        temperature=0.7,
+        max_tokens=11264,
+        temperature=0.01,
         top_p=1,
     )
     resp_text = response.choices[0].message.content

app_api.py

@@ -15,15 +15,15 @@ client = OpenAI(
 def chat(user_input, max_new_tokens=2048):
     # chat history
     messages_template = [
-        {"role": "system", "content": "You are Ling, an assistant created by inclusionAI"},
+ #       {"role": "system", "content": "You are Ling, an assistant created by inclusionAI"},
         {"role": "user", "content": user_input}
     ]
 
     response = client.chat.completions.create(
         model="Ling-lite-1.5-250604",
         messages=messages_template,
-        max_tokens=max_new_tokens,
-        temperature=0.7,
+        max_tokens=11264,
+        temperature=0.01,
         top_p=1,
     )
     yield response.choices[0].message.content