TypeError: Got unknown type content=

[Abe410]

I am running the following code snippet

from langchain.prompts.chat import (
    ChatPromptTemplate,
    HumanMessagePromptTemplate,
    SystemMessagePromptTemplate,
)
from langchain.schema import HumanMessage, SystemMessage
from langchain_openai import ChatOpenAI

chat = ChatOpenAI(temperature=0)

messages = [
    SystemMessage(
        content="Translate the sentence to English"
    ),
    HumanMessage(
        content="What is the price of Solana?"
    ),
]

print(chat(messages))

But I get the following error when I run it. I ran the exact same code yesterday but it was working fine. What can be the reason?

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[8], line 10
      1 messages = [
      2     SystemMessage(
      3         content="Translate the sentence to English"
   (...)
      7     ),
      8 ]
      9 start = time.time()
---> 10 print(chat(messages))
     11 total = time.time()-start

File /opt/conda/lib/python3.10/site-packages/langchain_core/_api/deprecation.py:145, in deprecated.<locals>.deprecate.<locals>.warning_emitting_wrapper(*args, **kwargs)
    143     warned = True
    144     emit_warning()
--> 145 return wrapped(*args, **kwargs)

File /opt/conda/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:690, in BaseChatModel.__call__(self, messages, stop, callbacks, **kwargs)
    682 @deprecated("0.1.7", alternative="invoke", removal="0.2.0")
    683 def __call__(
    684     self,
   (...)
    688     **kwargs: Any,
    689 ) -> BaseMessage:
--> 690     generation = self.generate(
    691         [messages], stop=stop, callbacks=callbacks, **kwargs
    692     ).generations[0][0]
    693     if isinstance(generation, ChatGeneration):
    694         return generation.message

File /opt/conda/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:407, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, **kwargs)
    405         if run_managers:
    406             run_managers[i].on_llm_error(e, response=LLMResult(generations=[]))
--> 407         raise e
    408 flattened_outputs = [
    409     LLMResult(generations=[res.generations], llm_output=res.llm_output)
    410     for res in results
    411 ]
    412 llm_output = self._combine_llm_outputs([res.llm_output for res in results])

File /opt/conda/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:397, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, **kwargs)
    394 for i, m in enumerate(messages):
    395     try:
    396         results.append(
--> 397             self._generate_with_cache(
    398                 m,
    399                 stop=stop,
    400                 run_manager=run_managers[i] if run_managers else None,
    401                 **kwargs,
    402             )
    403         )
    404     except BaseException as e:
    405         if run_managers:

File /opt/conda/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:576, in BaseChatModel._generate_with_cache(self, messages, stop, run_manager, **kwargs)
    572     raise ValueError(
    573         "Asked to cache, but no cache found at `langchain.cache`."
    574     )
    575 if new_arg_supported:
--> 576     return self._generate(
    577         messages, stop=stop, run_manager=run_manager, **kwargs
    578     )
    579 else:
    580     return self._generate(messages, stop=stop, **kwargs)

File /opt/conda/lib/python3.10/site-packages/langchain_openai/chat_models/base.py:431, in ChatOpenAI._generate(self, messages, stop, run_manager, stream, **kwargs)
    427     stream_iter = self._stream(
    428         messages, stop=stop, run_manager=run_manager, **kwargs
    429     )
    430     return generate_from_stream(stream_iter)
--> 431 message_dicts, params = self._create_message_dicts(messages, stop)
    432 params = {
    433     **params,
    434     **({"stream": stream} if stream is not None else {}),
    435     **kwargs,
    436 }
    437 response = self.client.create(messages=message_dicts, **params)

File /opt/conda/lib/python3.10/site-packages/langchain_openai/chat_models/base.py:448, in ChatOpenAI._create_message_dicts(self, messages, stop)
    446         raise ValueError("`stop` found in both the input and default params.")
    447     params["stop"] = stop
--> 448 message_dicts = [_convert_message_to_dict(m) for m in messages]
    449 return message_dicts, params

File /opt/conda/lib/python3.10/site-packages/langchain_openai/chat_models/base.py:448, in <listcomp>(.0)
    446         raise ValueError("`stop` found in both the input and default params.")
    447     params["stop"] = stop
--> 448 message_dicts = [_convert_message_to_dict(m) for m in messages]
    449 return message_dicts, params

File /opt/conda/lib/python3.10/site-packages/langchain_openai/chat_models/base.py:144, in _convert_message_to_dict(message)
    138     message_dict = {
    139         "role": "tool",
    140         "content": message.content,
    141         "tool_call_id": message.tool_call_id,
    142     }
    143 else:
--> 144     raise TypeError(f"Got unknown type {message}")
    145 if "name" in message.additional_kwargs:
    146     message_dict["name"] = message.additional_kwargs["name"]

TypeError: Got unknown type content='Translate the sentence to English'

[dosubot[bot]]

🤖

Hello @Abe410! Good to see you here again. Let's dive into this issue together.

Based on the error message you provided, it seems like the LangChain framework is unable to recognize the type of message you're passing to the ChatOpenAI instance. The error is raised at the point where the framework tries to convert your message into a dictionary format that it can process.

The error message TypeError: Got unknown type content='Translate the sentence to English' suggests that the SystemMessage and HumanMessage instances you're passing to the ChatOpenAI instance are not being recognized as valid input.

This could be due to a few reasons:

1. There might have been changes in the LangChain framework that modified how messages are processed. This could explain why your code was working fine yesterday but is throwing an error today.

2. The SystemMessage and HumanMessage classes might have been updated or deprecated, causing them to be unrecognized by the ChatOpenAI instance.

To troubleshoot this issue, you could try the following steps:

1. Check the version of LangChain you're currently using. You mentioned that the latest version is v0.1.0. If you're using an older version, consider updating to the latest version.

2. Review the documentation for the SystemMessage and HumanMessage classes to ensure you're using them correctly. Make sure the content attribute is being set correctly.

3. Look at the source code of the ChatOpenAI class, specifically the _create_message_dicts method where the error is being raised. This could give you more insight into what the method is expecting as input.

If none of these steps resolve your issue, please provide more information about your environment (e.g., Python version, operating system, etc.) and any other relevant details that could help diagnose the problem.

As for the similar issues found in the LangChain repository, they seem to be slightly different from your case. The first issue was about the agenerate method expecting a list of list of messages, and the second issue was resolved by updating the LangChain version. However, it's always a good idea to keep your packages up-to-date, so you might want to try updating LangChain to the latest version if you haven't done so already.

I hope this helps! If you have any other questions or need further clarification, feel free to ask.

Source Code Reference

Sources

• ChatOpenAI.agenerate seems broken in 0.0.168?
• run() got an unexpected keyword argument 'name'
• libs/langchain/langchain/tools/python/init.py

---

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[eyurtsev]

What version of langchain are you on? The code you posted works for me. see if upgrading to a newer version works?

[Abe410]

Yup, updating to the latest version worked for me.

[moussabader]

can you tell me which version of langchain you have?

[hectordanielc]

Yes please, I'm facing tha same error with the last version.

[naarkhoo]

what turn out to be the problem in my scenario to get the same error

was importing the right stuff

from langchain_openai import ChatOpenAI
    from langchain.prompts.chat import SystemMessage, HumanMessage, AIMessage
    from langchain_core.prompts import (
        ChatPromptTemplate,
        FewShotChatMessagePromptTemplate,
    )

instead of

from langchain_openai.chat_models import ChatOpenAI
from langchain.prompts import ChatPromptTemplate, FewShotChatMessagePromptTemplate

we used to spend tons of time with tensorflow ...