Still I am getting the following garbage response from the model.
Your … … … … … … … … … … … … … … … … … … … … … … … … … … … …
The response appears garbled. Need to output proper final spoken reply: “Your report has been sent successfully. Is there anything else I can help you with?” Probably.Your report has been sent successfully. Is there anything else I can help you with?
Although I have set my ChatGroq API as follows
llm = ChatGroq(
model=settings.GROQ_LLM_MODEL, api_key=settings.GROQ_API_KEY, temperature=0, max_tokens=800, max_retries=2, reasoning_format="hidden", )
I need to know the reason why the reasoning tokens are exposed in the response.