Tool Calling

Overview

Many AI applications interact directly with humans. In these cases, it is approrpiate for models to respond in natural langague. But what about cases where we want a model to also interact directly with systems, such as databases or an API? These systems often have a particular input schema; for example, APIs frequentlyhave a required payload structure. This need motivates the concept of tool calling. You can use tool calling to request model responses that match a particular schema.

info

You will sometimes hear the term function calling. We use this term interchangeably with tool calling.

Key Concepts

(1) Tool Creation: The tool needs to be described to the model so that the model knows what arguments to provide when it decides to call the tool. TODO: @eyurtsev to elaborate. (2) Tool Binding: The tool needs to be connected to a model that supports tool calling. This gives the model awarenes of the tool and the associated input schema required by the tool. (3) Tool Calling: When appropriate, the model can decide to call a tool and ensure its response conforms to the tool's input schema. (4) Tool Execution: The tool can be executed using the arguments provided by the model.

Tool Creation

TODO: @eyurtsev to add links and summary of the conceptual guide.

Tool Binding

Many model providers support tool calling.

tip

See our model integration page for a list of providers that support tool calling.

The central concept to understand is that LangChain provides a standardized interface for connecting tools to models. The .bind_tools() method can be used to specify which tools are available for a model to call.

model_with_tools = model.bind_tools([tools_list])

As a specific example, let's take a function multiply and bind it as a tool to a model that supports tool calling.

def multiply(a: int, b: int) -> int:
    """Multiply a and b.

    Args:
        a: first int
        b: second int
    """
    return a * b

llm_with_tools = tool_calling_model.bind_tools([multiply])

Tool Calling

A key principle of tool calling is that the model decides when to use a tool based on the input's relevance. The model doesn't always need to call a tool. For example, given an unrelated input, the model would not call the tool:

result = llm_with_tools.invoke("Hello world!")

The result would be an AIMessage containing the model's response in natural language (e.g., "Hello!"). However, if we pass an input relevant to the tool, the model should choose to call it:

result = llm_with_tools.invoke("What is 2 multiplied by 3?")

As before, the output result will be an AIMessage. But, if the tool was called, result will have a tool_calls attribute. This attribute includes everything needed to execute the tool, including the tool name and input arguments:

result.tool_calls
{'name': 'multiply', 'args': {'a': 2, 'b': 3}, 'id': 'xxx', 'type': 'tool_call'}

For more details on usage, see our how-to guides!

Tool execution

TODO: @eyurtsev @vbarda let's discuss this. This should also link to tool execution in LangGraph.

Best practices

When designing tools to be used by a model, it is important to keep in mind that:

• Models that have explicit tool-calling APIs will be better at tool calling than non-fine-tuned models.
• Models will perform better if the tools have well-chosen names and descriptions.
• Simple, narrowly scoped tools are easier for models to use than complex tools.
• Asking the model to select from a large list of tools poses challenges for the model.