Update app/llm.py

app/llm.py

@@ -104,11 +104,12 @@ async def generate(gen:GenModel):#, user: schemas.BaseUser = fastapi.Depends(cur
     gen.seed = 42
     try:
         st = time()
-        output = llm_generate.create_chat_completion(
-            messages=[
-                {"role": "system", "content": gen.system},
-                {"role": "user", "content": gen.question},
-            ],
+        output = llm_generate.create_completion(
+            #messages=[
+               # {"role": "system", "content": gen.system},
+              #  {"role": "user", "content": gen.question},
+           # ],
+            gen.question,
             temperature = gen.temperature,
             seed= gen.seed,
             chat_format="llama-2",