LangSmith provides special rendering and processing for LLM traces. This includes pretty rendering for the list of messages, token counting (assuming token counts are not available from the model provider) and token-based cost calculation.
In order to make the most of LangSmith’s LLM trace processing, we recommend logging your LLM traces in one of the specified formats. If you don’t log your LLM traces in the suggested formats, you will still be able to log the data to LangSmith, but it may not be processed or rendered in expected ways.
The examples below use the traceable decorator/wrapper to log the model run (which is the recommended approach for Python and JS/TS). However, the same idea applies if you are using the RunTree or API directly.

​
Using LangChain OSS or LangSmith wrappers

If you are using LangChain OSS to call language models or LangSmith wrappers (OpenAI, Anthropic), then you’re all set! These approaches will automatically log traces in the correct format.

​
Tracing a Model with a Custom Input/Output Format

When tracing a custom model, follow the guidelines below to ensure your LLM traces are rendered properly and features such as token tracking and cost calculation work as expected.

​
Input Format

A Python dictionary or TypeScript object containing a "messages" key with a list of messages in LangChain, OpenAI (chat completions) or Anthropic format. The messages key must be in the top level of the input field.
  • Each message must contain the key "role" and "content".
    • "role": "system" | "user" | "assistant" | "tool"
    • "content": string
  • Messages with the "assistant" role may contain "tool_calls". These "tool_calls" may be in OpenAI format or LangChain’s format.
  • LangSmith may use additional parameters in this input dict that match OpenAI’s chat completion endpoint for rendering in the trace view, such as a list of available tools for the model to call.
Here are some examples: 
inputs = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "What's the weather like in San Francisco?"},
    {
        "role": "assistant",
        "content": "I need to check the weather for you.",
        "tool_calls": [
            {
                "id": "call_XXX",
                "type": "function",
                "function": {
                    "name": "get_weather",
                    "arguments": '{"location": "San Francisco"}'
                }
            }
        ]
    },
    {
        "role": "tool",
        "tool_call_id": "call_XXXX",
        "content": "<weather>."
    }
]

​
Output Format

The output is accepted in any of the following formats:
  • A dictionary/object that contains the key message with a value that is a message object with the keys role and content.
  • A dictionary/object that contains the key choices with a value that is a list of dictionaries/objects. Each dictionary/object must contain the key message, which maps to a message object with the keys role and content.
  • A tuple/array of two elements, where the first element is the role and the second element is the content.
  • A dictionary/object that contains the key role and content.
  • Similar to the input format, outputs may contain "tool_calls". These "tool_calls" may be in OpenAI format or LangChain’s format.
Here are some examples: 
from langsmith import traceable

@traceable(run_type="llm")
def chat_model_message(messages):
    # Your model logic here
    return {
        "message": {
            "role": "assistant",
            "content": "Sure, what time would you like to book the table for?"
        }
    }

# Usage
inputs = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "I'd like to book a table for two."}
]
chat_model_message(inputs)

​
Converting custom I/O formats into LangSmith compatible formats

If you’re using a custom input or output format, you can convert it to a LangSmith compatible format using process_inputs and process_outputs functions on the @traceable decorator. Note that these parameters are only available in the Python SDK. process_inputs and process_outputs accept functions that allow you to transform the inputs and outputs of a specific trace before they are logged to LangSmith. They have access to the trace’s inputs and outputs, and can return a new dictionary with the processed data. Here’s a boilerplate example of how to use process_inputs and process_outputs to convert a custom I/O format into a LangSmith compatible format: 
class OriginalInputs(BaseModel):
    """Your app's custom request shape"""

class OriginalOutputs(BaseModel):
    """Your app's custom response shape."""

class LangSmithInputs(BaseModel):
    """The input format LangSmith expects."""

class LangSmithOutputs(BaseModel):
    """The output format LangSmith expects."""

def process_inputs(inputs: dict) -> dict:
    """Dict -> OriginalInputs -> LangSmithInputs -> dict"""

def process_outputs(output: Any) -> dict:
    """OriginalOutputs -> LangSmithOutputs -> dict"""


@traceable(run_type="llm", process_inputs=process_inputs, process_outputs=process_outputs)
def chat_model(inputs: dict) -> dict:
    """
    Your app's model call. Keeps your custom I/O shape.
    The decorators call process_* to log LangSmith-compatible format.
    """

​
Identifying a custom model in traces

When using a custom model, it is recommended to also provide the following metadata fields to identify the model when viewing traces and when filtering.
  • ls_provider: The provider of the model, eg “openai”, “anthropic”, etc.
  • ls_model_name: The name of the model, eg “gpt-4o-mini”, “claude-3-opus-20240307”, etc.
from langsmith import traceable

inputs = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "I'd like to book a table for two."},
]
output = {
    "choices": [
        {
            "message": {
                "role": "assistant",
                "content": "Sure, what time would you like to book the table for?"
            }
        }
    ]
}

@traceable(
    run_type="llm",
    metadata={"ls_provider": "my_provider", "ls_model_name": "my_model"}
)
def chat_model(messages: list):
    return output

chat_model(inputs)
The above code will log the following trace:
LangSmith UI showing an LLM call trace called ChatOpenAI with a system and human input followed by an AI Output.
If you implement a custom streaming chat_model, you can “reduce” the outputs into the same format as the non-streaming version. This is currently only supported in Python. 
def _reduce_chunks(chunks: list):
    all_text = "".join([chunk["choices"][0]["message"]["content"] for chunk in chunks])
    return {"choices": [{"message": {"content": all_text, "role": "assistant"}}]}

@traceable(
    run_type="llm",
    reduce_fn=_reduce_chunks,
    metadata={"ls_provider": "my_provider", "ls_model_name": "my_model"}
)
def my_streaming_chat_model(messages: list):
    for chunk in ["Hello, " + messages[1]["content"]]:
        yield {
            "choices": [
                {
                    "message": {
                        "content": chunk,
                        "role": "assistant",
                    }
                }
            ]
        }

list(
    my_streaming_chat_model(
        [
            {"role": "system", "content": "You are a helpful assistant. Please greet the user."},
            {"role": "user", "content": "polly the parrot"},
        ],
    )
)
If ls_model_name is not present in extra.metadata, other fields might be used from the extra.metadata for estimating token counts. The following fields are used in the order of precedence:
  1. metadata.ls_model_name
  2. inputs.model
  3. inputs.model_name
To learn more about how to use the metadata fields, see this guide.

​
Provide token and cost information

By default, LangSmith uses tiktoken to count tokens, utilizing a best guess at the model’s tokenizer based on the ls_model_name provided. It also calculates costs automatically by using the model pricing table. To learn how LangSmith calculates token-based costs, see this guide. However, many models already include exact token counts as part of the response. If you have this information, you can override the default token calculation in LangSmith in one of two ways:
  1. Extract usage within your traced function and set a usage_metadata field on the run’s metadata.
  2. Return a usage_metadata field in your traced function outputs.
In both cases, the usage metadata you send should contain a subset of the following LangSmith-recognized fields:
You cannot set any fields other than the ones listed below. You do not need to include all fields.
class UsageMetadata(TypedDict, total=False):
    input_tokens: int
    """The number of tokens used for the prompt."""
    output_tokens: int
    """The number of tokens generated as output."""
    total_tokens: int
    """The total number of tokens used."""
    input_token_details: dict[str, float]
    """The details of the input tokens."""
    output_token_details: dict[str, float]
    """The details of the output tokens."""
    input_cost: float
    """The cost of the input tokens."""
    output_cost: float
    """The cost of the output tokens."""
    total_cost: float
    """The total cost of the tokens."""
    input_cost_details: dict[str, float]
    """The cost details of the input tokens."""
    output_cost_details: dict[str, float]
    """The cost details of the output tokens."""
Note that the usage data can also include cost information, in case you do not want to rely on LangSmith’s token-based cost formula. This is useful for models with pricing that is not linear by token type.

​
Setting run metadata

You can modify the current run’s metadata with usage information within your traced function. The advantage of this approach is that you do not need to change your traced function’s runtime outputs. Here’s an example:
Requires langsmith>=0.3.43 (Python) and langsmith>=0.3.30 (JS/TS).
from langsmith import traceable, get_current_run_tree

inputs = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "I'd like to book a table for two."},
]

@traceable(
    run_type="llm",
    metadata={"ls_provider": "my_provider", "ls_model_name": "my_model"}
)
def chat_model(messages: list):
    llm_output = {
        "choices": [
            {
                "message": {
                    "role": "assistant",
                    "content": "Sure, what time would you like to book the table for?"
                }
            }
        ],
        "usage_metadata": {
            "input_tokens": 27,
            "output_tokens": 13,
            "total_tokens": 40,
            "input_token_details": {"cache_read": 10},
            # If you wanted to specify costs:
            # "input_cost": 1.1e-6,
            # "input_cost_details": {"cache_read": 2.3e-7},
            # "output_cost": 5.0e-6,
        },
    }
    run = get_current_run_tree()
    run.set(usage_metadata=llm_output["usage_metadata"])
    return llm_output["choices"][0]["message"]

chat_model(inputs)

​
Setting run outputs

You can add a usage_metadata key to the function’s response to set manual token counts and costs. 
from langsmith import traceable

inputs = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "I'd like to book a table for two."},
]
output = {
    "choices": [
        {
            "message": {
                "role": "assistant",
                "content": "Sure, what time would you like to book the table for?"
            }
        }
    ],
    "usage_metadata": {
        "input_tokens": 27,
        "output_tokens": 13,
        "total_tokens": 40,
        "input_token_details": {"cache_read": 10},
        # If you wanted to specify costs:
        # "input_cost": 1.1e-6,
        # "input_cost_details": {"cache_read": 2.3e-7},
        # "output_cost": 5.0e-6,
    },
}

@traceable(
    run_type="llm",
    metadata={"ls_provider": "my_provider", "ls_model_name": "my_model"}
)
def chat_model(messages: list):
    return output

chat_model(inputs)

​
Time-to-first-token

If you are using traceable or one of our SDK wrappers, LangSmith will automatically populate time-to-first-token for streaming LLM runs. However, if you are using the RunTree API directly, you will need to add a new_token event to the run tree in order to properly populate time-to-first-token. Here’s an example: 
from langsmith.run_trees import RunTree
run_tree = RunTree(
    name="CustomChatModel",
    run_type="llm",
    inputs={ ... }
)
run_tree.post()
llm_stream = ...
first_token = None
for token in llm_stream:
    if first_token is None:
      first_token = token
      run_tree.add_event({
        "name": "new_token"
      })
run_tree.end(outputs={ ... })
run_tree.patch()