minor fix

routers/get_chatrespone.py

@@ -28,10 +28,9 @@ async def get_chat_respone(body: ChatInputForm, api_key: str = Depends(get_api_k
 
     prompt = get_prompt(body.prompt)
 
-    promptTemplate = PromptTemplate.from_template(prompt)
     try:
         llm = OllamaLLM(
-            model="llama3.2",
+            model=body.repo_id,
             temperature=0.2,
             # huggingfacehub_api_token=HUGGINGFACEHUB_API_TOKEN,
         )
@@ -41,8 +40,6 @@ async def get_chat_respone(body: ChatInputForm, api_key: str = Depends(get_api_k
             ("human", body.textInput)
         ]
 
-
-        llm_chain = promptTemplate | llm
         response = llm.stream(messages)
 
         return StreamingResponse(get_response(response), media_type='text/event-stream')