AttributeError: 'VertexAI' object has no attribute 'bind_tools'

[jaymon0703]

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

from langchain_core.tools import tool
from langchain_community.llms import VertexAI
from google.cloud import aiplatform

aiplatform.init(
    # your Google Cloud Project ID or number
    # environment default used is not set
    project='your_project_id',
    location="your_preferred_location"
)

model = VertexAI(model_name="gemini-pro", seed=123, max_output_tokens=8192, temperature=0)


@tool
def magic_function(input: int) -> int:
    """Applies a magic function to an input."""
    return input + 2


tools = [magic_function]


query = "what is the value of magic_function(3)?"

from langgraph.prebuilt import create_react_agent

app = create_react_agent(model, tools)


messages = app.invoke({"messages": [("human", query)]})
{
    "input": query,
    "output": messages["messages"][-1].content,
}

Description

When i run the above code, which is a slight modification of the LangGraph example code for migrating from Agent Executor to LangGraph (https://python.langchain.com/v0.2/docs/how_to/migrate_agent/) i get the error:

AttributeError: 'VertexAI' object has no attribute 'bind_tools'

Is there a way around this?

System Info

System Information

OS: Windows
OS Version: 10.0.19045
Python Version: 3.10.0 (tags/v3.10.0:b494f59, Oct 4 2021, 19:00:18) [MSC v.1929 64 bit (AMD64)]

Package Information

langchain_core: 0.2.23
langchain: 0.2.11
langchain_community: 0.0.36
langsmith: 0.1.93
langchain_experimental: 0.0.49
langchain_google_vertexai: 0.0.2
langchain_openai: 0.0.8
langchain_text_splitters: 0.2.2
langchainhub: 0.1.14
langgraph: 0.1.14

[dosubot[bot]]

The VertexAI model does not have a bind_tools method. However, you can still use VertexAI with tools and agents in LangChain by following the correct setup. Here's how you can modify your code to work with VertexAI:

1. Ensure you have the correct imports and setup for VertexAI.
2. Use the create_react_agent function correctly without relying on bind_tools.

Here's the updated code:

from langchain_core.tools import tool
from langchain_community.llms import VertexAI
from google.cloud import aiplatform

# Initialize Google Cloud AI Platform
aiplatform.init(
    project='your_project_id',
    location="your_preferred_location"
)

# Create a VertexAI model instance
model = VertexAI(model_name="gemini-pro", seed=123, max_output_tokens=8192, temperature=0)

# Define a tool using the tool decorator
@tool
def magic_function(input: int) -> int:
    """Applies a magic function to an input."""
    return input + 2

# List of tools
tools = [magic_function]

# Query to be processed
query = "what is the value of magic_function(3)?"

# Import the create_react_agent function
from langgraph.prebuilt import create_react_agent

# Create the agent using the model and tools
app = create_react_agent(model, tools)

# Invoke the agent with the query
messages = app.invoke({"messages": [("human", query)]})

# Print the output
print({
    "input": query,
    "output": messages["messages"][-1].content,
})

This setup ensures that VertexAI is used correctly with the tools and agent creation process in LangChain. The create_react_agent function does not require bind_tools and will work with the provided tools and model ^[1][2].

To continue talking to Dosu, mention @dosu.

[jaymon0703]

So ChatVertexAI from langchain_google_vertexai lib has a bind_tools method...using that instead