| else: | else: | ||||
| prompt_messages.append(AssistantPromptMessage(content=message.answer)) | prompt_messages.append(AssistantPromptMessage(content=message.answer)) | ||||
| if not prompt_messages: | |||||
| return [] | |||||
| if not prompt_messages: | |||||
| return [] | |||||
| # prune the chat message if it exceeds the max token limit | |||||
| curr_message_tokens = self.model_instance.get_llm_num_tokens(prompt_messages) | |||||
| # prune the chat message if it exceeds the max token limit | |||||
| curr_message_tokens = self.model_instance.get_llm_num_tokens(prompt_messages) | |||||
| if curr_message_tokens > max_token_limit: | if curr_message_tokens > max_token_limit: | ||||
| while curr_message_tokens > max_token_limit and len(prompt_messages) > 1: | while curr_message_tokens > max_token_limit and len(prompt_messages) > 1: |