WIP: tunning

ai_service/main.py

@@ -32,12 +32,15 @@ async def message(
 
     # Build prompt (very simple example)
     prompt = f"User {payload.userId} said: {payload.content}\nBot:"
-    resp = openai.Completion.create(
-        model="text-davinci-003",
-        prompt=prompt,
-        max_tokens=150
+    chat_response = opeai.ChatCompletion.create(
+        model="gpt-3.5-turbo",
+        messages=[
+            {"role": "system", "content": "-"},
+            {"role": "user", "content": prompt}
+        ]
+        temperature=0.7,
     )
-    reply = resp.choices[0].text.strip()
+    reply = chat_response.choices[0].text.strip()
 
     # Cache reply for idempotency
     r.set(payload.eventId, reply, ex=3600)