---
id: langchainplus-sdk
version: "0.0.20"
license: MIT
license_treatment: permissive
maintenance: abandoned
---

# langchainplus-sdk — Client library to connect to the LangSmith LLM Tracing and Evaluation Platform.

License: permissive · Maintenance: abandoned · Downloads: 702.3K/mo

## What it is and what it does

The langchainplus-sdk is a Python client for LangSmith, a platform designed to help teams develop, debug, and evaluate language model applications. It provides two main integration paths: seamless tracing within LangChain applications via environment variables, and standalone trace logging through RunTree objects for non-LangChain frameworks. The package depends on pydantic for data validation, requests for HTTP communication, and tenacity for retry logic.

Typical workflows involve setting up LangSmith credentials, logging execution traces from your LLM application, then using the platform to create datasets from those traces and run evaluations. The SDK handles batching and posting of nested run hierarchies to the LangSmith backend. However, the package has not been updated since July 2023 and is marked as abandoned, meaning no active maintenance, bug fixes, or compatibility updates are being provided.

Use it for:

- Log traces from LangChain applications by setting environment variables and letting the SDK automatically capture execution flow.
- Track custom LLM, chain, and tool invocations outside LangChain using RunTree to build hierarchical execution records.
- Convert stored execution runs into datasets for model evaluation and fine-tuning experiments.
- Evaluate LLM outputs against reference answers using custom scoring functions via the StringEvaluator interface.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Client library for connecting to LangSmith, a platform for tracing, debugging, and evaluating language model applications and agents.

No—this package is abandoned (last release July 2023, 1138 days ago) with no active maintenance. If you need LangSmith integration, verify whether LangChain itself now includes native support or whether LangSmith has released a newer, actively maintained client. Installing an unmaintained SDK for a third-party platform creates long-term maintenance and compatibility risk.

## Install

pip install langchainplus-sdk
uv add langchainplus-sdk
poetry add langchainplus-sdk

## Installing langchainplus-sdk

Before you install: Low install friction with three common runtime dependencies (pydantic, requests, tenacity). However, the package is abandoned—last release was 2023-07-03, over 1138 days ago. No active maintenance or security updates.

License in practice: MIT license permits commercial and private use with minimal restrictions, requiring only license and copyright notice retention.

Quickstart:

pip install langchainplus-sdk

from langchainplus_sdk.run_trees import RunTree

run = RunTree(
    name="My Agent",
    run_type="chain",
    inputs={"text": "example"},
    serialized={}
)
run.end(outputs={"result": "done"})
run.post()

Requires LANGCHAIN_ENDPOINT and LANGCHAIN_API_KEY environment variables or explicit connection parameters to connect to LangSmith.

Verify before relying:

- Whether LangSmith platform itself remains actively maintained and whether this SDK version remains compatible with current platform APIs.
- Whether the package has been superseded by a newer client library or integration method.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: abandoned
- Downloads: 702.3K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

langsmith tracing client, llm application monitoring, language model debugging, agent evaluation platform, trace logging for llms, langchain integration client, run tree tracking, llm-tracing, abandoned

[View on SkillFed](https://skillfed.io/packages/langchainplus-sdk) · [View on PyPI](https://pypi.org/project/langchainplus-sdk/)