NotImplementedError: get_num_tokens_from_messages() is not presently implemented for model cl100k_base

[zzc2004]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

import os

# 导入所需的库
from langchain_openai import ChatOpenAI
from langchain.chains import ConversationChain
from langchain.memory import  ConversationSummaryBufferMemory


# 初始化大语言模型
llm = ChatOpenAI(
    temperature=0.5,
    model=os.environ.get("LLM_MODELEND"),
)

# 初始化对话链
conversation = ConversationChain(
    llm=llm, memory=ConversationSummaryBufferMemory(llm=llm, max_token_limit=300)
)

# 第一天的对话
# 回合1
result = conversation("我姐姐明天要过生日，我需要一束生日花束。")
print(result)
# 回合2
result = conversation("她喜欢粉色玫瑰，颜色是粉色的。")
# print("\n第二次对话后的记忆:\n", conversation.memory.buffer)
print(result)

# 第二天的对话
# 回合3
result = conversation("我又来了，还记得我昨天为什么要来买花吗？")
print(result)

Description

我正在用langchain中的ConversationSummaryBufferMemory学习langchain的记忆功能，但是在运行后会有以下错误提示，NotImplementedError: get_num_tokens_from_messages() is not presently implemented for model cl100k_base

System Info

python 3.11，windows11，langchain0.3

[lms2004]

ChartOpenAI Implementation

# ChatOpenAI:
#     def get_num_tokens_from_messages(self, messages: List[BaseMessage]) -> int:

#         if sys.version_info[1] <= 7:
#             return super().get_num_tokens_from_messages(messages)
#         model, encoding = self._get_encoding_model()
#         if model.startswith("gpt-3.5-turbo-0301"):
#             # every message follows <im_start>{role/name}\n{content}<im_end>\n
#             tokens_per_message = 4
#             # if there's a name, the role is omitted
#             tokens_per_name = -1
#         elif model.startswith("gpt-3.5-turbo") or model.startswith("gpt-4"):
#             tokens_per_message = 3
#             tokens_per_name = 1
#         else:
#             raise NotImplementedError(
#                 f"get_num_tokens_from_messages() is not presently implemented "
#                 f"for model {model}. See "
#                 "https://platform.openai.com/docs/guides/text-generation/managing-tokens"  # noqa: E501
#                 " for information on how messages are converted to tokens."
#             )


#     def _get_encoding_model(self) -> Tuple[str, tiktoken.Encoding]:
#         if self.tiktoken_model_name is not None:
#             model = self.tiktoken_model_name
#         else:
#             model = self.model_name
#         try:
#             encoding = tiktoken.encoding_for_model(model)
#         except KeyError:
#             model = "cl100k_base"
#             encoding = tiktoken.get_encoding(model)
#         return model, encoding
# """

---

UPDATE: Updated Your Code

class ChatOpenAIIn05(ChatOpenAI):
    def _get_encoding_model(self) -> Tuple[str, tiktoken.Encoding]:
        """
        Override the method to return a hardcoded valid model and its encoding.
        """
        # Set the model to a valid one to avoid errors
        model = "gpt-3.5-turbo"
        return model, tiktoken.encoding_for_model(model)


# Updated code (avoids errors)
llm = ChatOpenAIIn05(
    temperature=0.5,
    model=os.environ.get("LLM_MODELEND"),  # Environment variable is still used
)

---