Agent react cannot accept gpt4 or gpt 4o as input model

[HelloWorldLTY]

Checked other resources

• [X] I added a very descriptive title to this issue.
• [X] I searched the LangChain documentation with the integrated search.
• [X] I used the GitHub search to find a similar question and didn't find it.
• [X] I am sure that this is a bug in LangChain rather than my code.
• [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).

Example Code

The mimial reproduce code is:

from langchain import hub
from langchain.agents import AgentExecutor, create_react_agent
from langchain_community.tools.tavily_search import TavilySearchResults
from langchain_openai import OpenAI

tools = []

# Get the prompt to use - you can modify this!
prompt = hub.pull("hwchase17/react")

# Choose the LLM to use
llm = ChatOpenAI(temperature=0, model="gpt-4o")

# Construct the ReAct agent
agent = create_react_agent(llm, tools, prompt)



# Create an agent executor by passing in the agent and tools
agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)

agent_executor.invoke({"input": "what is LangChain?"})

Error Message and Stack Trace (if applicable)

ile /home/tl688/.conda/envs/cell2sentence/lib/python3.10/site-packages/langchain/agents/output_parsers/react_single_input.py:75, in ReActSingleInputOutputParser.parse(self, text) 74 if not re.search(r"Action\s*\d*\s*:[\s](.?)", text, re.DOTALL): ---> 75 raise OutputParserException( 76 f"Could not parse LLM output: {text}", 77 observation=MISSING_ACTION_AFTER_THOUGHT_ERROR_MESSAGE, 78 llm_output=text, 79 send_to_llm=True, 80 ) 81 elif not re.search( 82 r"[\s]Action\s\d*\sInput\s\d*\s*:[\s](.)", text, re.DOTALL 83 ):

OutputParserException: Could not parse LLM output: LangChain is a framework designed to facilitate the development of applications powered by language models. It provides a suite of tools and components that help developers build applications that can interact with language models in a structured and efficient manner. LangChain is particularly useful for creating applications that require complex language processing tasks, such as chatbots, virtual assistants, and other AI-driven communication tools. It supports integration with various language models and offers features for managing conversations, handling inputs and outputs, and maintaining context across interactions. For troubleshooting, visit: https://python.langchain.com/docs/troubleshooting/errors/OUTPUT_PARSING_FAILURE

Description

This error only happended in gpt4 and gpt4o model, not in 3.5. Replacing it with OpenAI can work.

System Info

The updated version of langchain.

[keenborder786]

@HelloWorldLTY, which version of langchain are you using. Can you uprade to the latest version. Since I tested the same code and everything seems to be working.

[keenborder786]

But @HelloWorldLTY if you are facing this error then do the following:

agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True, handle_parsing_errors=True)

Making handle_parsing_errors=True will make sure that agent executor takes care of any parsing error by feeding the parsing error back to llm.

[HelloWorldLTY]

Hi, I have tried handle_parsing_errors=True but it will repeat error information for me. I think I have updated my longchain. Here is my info:

kiwisolver==1.4.7 langchain==0.3.7 langchain-anthropic==0.3.0 langchain-community==0.3.5 langchain-core==0.3.24 langchain-experimental==0.3.3 langchain-huggingface==0.1.2 langchain-ollama==0.2.0 langchain-openai==0.2.5 langchain-text-splitters==0.3.2 langgraph==0.2.44 langgraph-checkpoint==2.0.2 langgraph-sdk==0.1.35 langsmith==0.1.139

[HelloWorldLTY]

Hi, does it mean I can try reactgpt4 or gpt4o with the referred comments? Thanks.

[dosubot[bot]]

Hi, @HelloWorldLTY. I'm Dosu, and I'm helping the LangChain team manage their backlog. I'm marking this issue as stale.

Issue Summary:

• You reported a bug where the ReAct agent cannot process "gpt-4o" as an input model, resulting in an OutputParserException.
• Keenborder786 suggested upgrading LangChain and using handle_parsing_errors=True.
• You noted that the suggested solution only repeats the error information and provided your package versions.
• You are seeking clarification on using "reactgpt4" or "gpt4o" with the suggested adjustments.

Next Steps:

• Could you confirm if this issue is still relevant with the latest version of LangChain? If so, please comment to keep the discussion open.
• If there is no further activity, this issue will be automatically closed in 7 days.

Thank you for your understanding and contribution!