LangChain LCEL - TypeError: Serializable.__init__() takes 1 positional argument but 2 were given

Question

When I run:

from langchain.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser

model = ChatOpenAI()
prompt = ChatPromptTemplate.from_template("tell me a joke about {topic}")
chain = prompt | model | StrOutputParser
chain.invoke({"topic": "bears"})

I get:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
[<ipython-input-48-da226a1d50ac>](https://localhost:8080/#) in <cell line: 8>()
      6 prompt = ChatPromptTemplate.from_template("tell me a joke about {topic}")
      7 chain = prompt | model | StrOutputParser
----> 8 chain.invoke({"topic": "bears"})

5 frames
[/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/base.py](https://localhost:8080/#) in invoke(self, input, config)
   1512         try:
   1513             for i, step in enumerate(self.steps):
-> 1514                 input = step.invoke(
   1515                     input,
   1516                     # mark each step as a child run

[/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/base.py](https://localhost:8080/#) in invoke(self, input, config, **kwargs)
   2629         """Invoke this runnable synchronously."""
   2630         if hasattr(self, "func"):
-> 2631             return self._call_with_config(
   2632                 self._invoke,
   2633                 input,

[/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/base.py](https://localhost:8080/#) in _call_with_config(self, func, input, config, run_type, **kwargs)
    884         )
    885         try:
--> 886             output = call_func_with_variable_args(
    887                 func, input, config, run_manager, **kwargs
    888             )

[/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/config.py](https://localhost:8080/#) in call_func_with_variable_args(func, input, config, run_manager, **kwargs)
    306     if run_manager is not None and accepts_run_manager(func):
    307         kwargs["run_manager"] = run_manager
--> 308     return func(input, **kwargs)  # type: ignore[call-arg]
    309 
    310 

[/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/base.py](https://localhost:8080/#) in _invoke(self, input, run_manager, config, **kwargs)
   2547         **kwargs: Any,
   2548     ) -> Output:
-> 2549         output = call_func_with_variable_args(
   2550             self.func, input, config, run_manager, **kwargs
   2551         )

[/usr/local/lib/python3.10/dist-packages/langchain_core/runnables/config.py](https://localhost:8080/#) in call_func_with_variable_args(func, input, config, run_manager, **kwargs)
    306     if run_manager is not None and accepts_run_manager(func):
    307         kwargs["run_manager"] = run_manager
--> 308     return func(input, **kwargs)  # type: ignore[call-arg]
    309 
    310 

TypeError: Serializable.__init__() takes 1 positional argument but 2 were given

I even get the same thing from this very simple example:

prompt2 = ChatPromptTemplate.from_template("What's up?")
chain = prompt2 | model | StrOutputParser
chain.invoke({})

I've followed the LangChain docs on Prompt + LLM, and it looks like I'm doing the same thing as the official docs.

Answer 1

Output parsers require parentheses. So, instead of StrOutputParser, call StrOutputParser().

Try this:

from langchain.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser

model = ChatOpenAI()
prompt = ChatPromptTemplate.from_template("tell me a joke about {topic}")
chain = prompt | model | StrOutputParser()
chain.invoke({"topic": "bears"})