Skip to main content
The AsyncClient class provides an async interface for interacting with the LangSmith API. It uses httpx for async HTTP requests and is ideal for async applications.

​
Constructor

from langsmith import AsyncClient

client = AsyncClient(
    api_url="https://api.smith.langchain.com",
    api_key="your-api-key"
)
​
api_url
str | None
URL for the LangSmith API. Defaults to the LANGCHAIN_ENDPOINT or LANGSMITH_ENDPOINT environment variable.
​
api_key
str | None
API key for authentication. Defaults to the LANGCHAIN_API_KEY or LANGSMITH_API_KEY environment variable.
​
timeout_ms
int | tuple[int, int, int, int] | None
Timeout for requests in milliseconds. Can be a single value or a tuple of timeouts.
​
retry_config
Mapping[str, Any] | None
Configuration for request retries. Default is {"max_retries": 3}.
​
web_url
str | None
URL for the LangSmith web application.
​
disable_prompt_cache
bool
Whether to disable prompt caching for this client. Default is False.

​
Usage with context manager

The async client should be used as an async context manager: 
async with AsyncClient() as client:
    await client.acreate_run(
        name="my-chain",
        inputs={"query": "What is LangSmith?"},
        run_type="chain"
    )

​
Core methods

​
acreate_run

Asynchronously create a new run in LangSmith. 
await client.acreate_run(
    name="my-chain",
    inputs={"query": "What is LangSmith?"},
    run_type="chain",
    project_name="my-project"
)
​
name
str
required
Name of the run.
​
inputs
dict[str, Any]
required
Input data for the run.
​
run_type
str
required
Type of run: "llm", "chain", "tool", "retriever", or "prompt".
​
project_name
str | None
Project to log the run to.
​
run_id
UUID | str | None
Unique identifier for the run.
​
parent_run_id
UUID | str | None
ID of the parent run for nested traces.
​
tags
list[str] | None
Tags for filtering and organization.
​
metadata
dict[str, Any] | None
Additional metadata.

​
aupdate_run

Asynchronously update an existing run. 
await client.aupdate_run(
    run_id=run_id,
    outputs={"answer": "LangSmith is a platform for..."},
    end_time=datetime.now()
)
​
run_id
UUID | str
required
ID of the run to update.
​
outputs
dict[str, Any] | None
Output data from the run.
​
error
str | None
Error message if the run failed.
​
end_time
datetime | None
End time of the run.

​
acreate_feedback

Asynchronously create feedback for a run. 
await client.acreate_feedback(
    run_id=run_id,
    key="accuracy",
    score=0.95
)
​
run_id
UUID | str
required
ID of the run to attach feedback to.
​
key
str
required
Name of the metric.
​
score
float | int | bool | None
Numeric score.
​
comment
str | None
Explanation for the feedback.

​
Dataset methods

​
acreate_dataset

Asynchronously create a new dataset. 
dataset = await client.acreate_dataset(
    dataset_name="my-dataset",
    description="QA pairs for evaluation"
)
​
dataset_name
str
required
Name of the dataset.
​
description
str | None
Description of the dataset.
​
return
Dataset
The created dataset object.

​
acreate_example

Asynchronously add an example to a dataset. 
example = await client.acreate_example(
    inputs={"question": "What is LangSmith?"},
    outputs={"answer": "A platform for LLM development"},
    dataset_id=dataset.id
)
​
inputs
dict[str, Any]
required
Input data for the example.
​
outputs
dict[str, Any] | None
Expected output data.
​
dataset_id
UUID | str
required
ID of the dataset.
​
return
Example
The created example object.

​
alist_datasets

Asynchronously list all datasets. 
async for dataset in client.alist_datasets():
    print(dataset.name)
​
return
AsyncIterator[Dataset]
Async iterator of dataset objects.

​
Prompt methods

​
apull_prompt

Asynchronously pull a prompt from the hub. 
prompt = await client.apull_prompt("my-prompt")
​
prompt_identifier
str
required
Prompt identifier in format "name", "owner/name", or with commit hash.
​
include_model
bool | None
Whether to include model configuration.
​
return
Any
The prompt object.

​
apush_prompt

Asynchronously push a prompt to the hub. 
await client.apush_prompt(
    "my-prompt",
    object=prompt_template
)
​
prompt_identifier
str
required
Name for the prompt.
​
object
Any
required
The prompt object to push.
​
return
str
URL of the pushed prompt.

​
Cleanup

Always close the async client when done: 
# Using context manager (recommended)
async with AsyncClient() as client:
    # Use client
    pass

# Or manually
client = AsyncClient()
try:
    # Use client
    pass
finally:
    await client.aclose()

Build docs developers (and LLMs) love

Get started for freeTalk to us