Langfuse

https://langfuse.com/docs
Admin
Langfuse is an open-source LLM engineering platform that helps teams collaboratively debug, analyze,...
Tokens:201,633
Snippets:2,449
Trust Score:9.9
Update:1 month ago
Show doc for...
Tokens:
Raw
### Install and Start Flowise

Source: https://langfuse.com/docs/flowise

Commands to install the Flowise package globally and start the application server.

```bash
# install
npm install -g flowise

# start
npx flowise start
```

--------------------------------

### Install and Configure JS/TS SDK

Source: https://langfuse.com/docs/prompt-management/get-started

Install the client library and initialize the Langfuse client.

```bash
npm i @langfuse/client
```

```bash
LANGFUSE_SECRET_KEY = "sk-lf-..."
LANGFUSE_PUBLIC_KEY = "pk-lf-..."
LANGFUSE_BASE_URL = "https://cloud.langfuse.com" # 🇪🇺 EU region
# Other Langfuse data regions include 🇺🇸 US: https://us.cloud.langfuse.com, 🇯🇵 Japan: https://jp.cloud.langfuse.com and ⚕️ HIPAA: https://hipaa.cloud.langfuse.com
```

```ts
import { LangfuseClient } from "@langfuse/client";

const langfuse = new LangfuseClient();
```

--------------------------------

### Installing Langfuse and OpenAI SDK

Source: https://langfuse.com/docs/integrations/openai/python/get-started

Install the necessary packages via pip.

```bash
pip install langfuse openai
```

--------------------------------

### Install and initialize JS/TS SDK

Source: https://langfuse.com/docs/api-and-data-platform/features/query-via-sdk

Installation command and client initialization for the JS/TS SDK.

```bash
npm install @langfuse/client
```

```typescript
import { LangfuseClient } from "@langfuse/client";

const langfuse = new LangfuseClient();
```

--------------------------------

### Install DeepEval

Source: https://langfuse.com/docs/evaluation/features/synthetic-datasets

Install the DeepEval library.

```bash
%pip install deepeval
```

--------------------------------

### Install OpenTelemetry SDK

Source: https://langfuse.com/docs/get-started

Install the OpenTelemetry SDK required for data capture.

```bash
npm install @opentelemetry/sdk-node
```

--------------------------------

### Install dependencies

Source: https://langfuse.com/docs/opentelemetry/example-openlit

Install the required Python packages for OpenAI, Langfuse, and OpenLIT.

```bash
%pip install openai langfuse openlit --upgrade
```

--------------------------------

### Install and Initialize Python SDK

Source: https://langfuse.com/docs/api-and-data-platform/features/query-via-sdk

Install the package and initialize the client using environment variables for authentication.

```bash
pip install langfuse
```

```python
from langfuse import get_client
langfuse = get_client()  # uses environment variables to authenticate
```

--------------------------------

### Install and Configure Python SDK

Source: https://langfuse.com/docs/prompt-management/get-started

Install the package and set environment variables for authentication and region selection.

```bash
pip install langfuse
```

```bash
LANGFUSE_SECRET_KEY = "sk-lf-..."
LANGFUSE_PUBLIC_KEY = "pk-lf-..."
LANGFUSE_BASE_URL = "https://cloud.langfuse.com" # 🇪🇺 EU region
# Other Langfuse data regions include 🇺🇸 US: https://us.cloud.langfuse.com, 🇯🇵 Japan: https://jp.cloud.langfuse.com and ⚕️ HIPAA: https://hipaa.cloud.langfuse.com
```

--------------------------------

### Install Dependencies

Source: https://langfuse.com/docs/prompt-management/features/github-integration

Install the required Python packages for the FastAPI server.

```bash
pip install fastapi uvicorn pydantic-settings httpx
```

--------------------------------

### Install via Agent Instruction

Source: https://langfuse.com/docs/api-and-data-platform/features/agent-skill

Use this natural language prompt to instruct your coding agent to install the skill.

```txt
"Install the Langfuse Agent Skill from github.com/langfuse/skills."
```

--------------------------------

### Install Pi MCP Adapter

Source: https://langfuse.com/docs/api-and-data-platform/features/mcp-server

Install the community-maintained extension to enable MCP support in Pi.

```bash
pi install npm:pi-mcp-adapter
```

--------------------------------

### Install required security and tracing packages

Source: https://langfuse.com/docs/security-and-guardrails

Install the necessary libraries for LLM Guard, Langfuse, and OpenAI integration.

```bash
pip install llm-guard langfuse openai
```

--------------------------------

### Install Required Packages

Source: https://langfuse.com/docs/security/example-python

Install the necessary libraries for Langfuse and LLM Guard integration.

```bash
%pip install llm-guard langfuse openai --upgrade
```

--------------------------------

### Install Langfuse SDKs

Source: https://langfuse.com/docs/sdk/overview

Commands to install the necessary packages for Python and JS/TS environments.

```bash
pip install langfuse
```

```bash
npm install @langfuse/tracing @langfuse/otel @opentelemetry/sdk-node
```

--------------------------------

### Install LangChain dependencies

Source: https://langfuse.com/docs/get-started

Install the Langfuse and LangChain OpenAI packages.

```bash
pip install langfuse langchain-openai
```

--------------------------------

### Install Langfuse and OpenTelemetry dependencies

Source: https://langfuse.com/docs/get-started

Install the necessary packages for tracing and OpenTelemetry integration.

```bash
npm install @langfuse/tracing @langfuse/otel @opentelemetry/sdk-node
```

--------------------------------

### Install Dependencies

Source: https://langfuse.com/docs/opentelemetry/example-mlflow

Install the required Python packages for MLflow, OpenAI, and OpenTelemetry integration.

```bash
%pip install mlflow openai opentelemetry-exporter-otlp-proto-http langfuse --upgrade
```

--------------------------------

### Clone Example Document Dataset

Source: https://langfuse.com/docs/evaluation/features/synthetic-datasets

Download a sample markdown document set for testing.

```bash
# Clone an example document set
!git clone https://huggingface.co/datasets/explodinggradients/Sample_Docs_Markdown
```

--------------------------------

### Install Langfuse Browser SDK

Source: https://langfuse.com/docs/evaluation/evaluation-methods/custom-scores

Install the browser SDK package for frontend score ingestion.

```bash
npm install @langfuse/browser
```

--------------------------------

### Run Server Locally

Source: https://langfuse.com/docs/prompt-management/features/github-integration

Start the FastAPI server using uvicorn for local development.

```bash
uvicorn main:app --reload --port 8000
```

--------------------------------

### Install Langfuse SDK

Source: https://langfuse.com/docs/get-started

Install the Langfuse package via pip to enable tracing capabilities.

```bash
pip install langfuse
```

--------------------------------

### Install Vercel AI SDK dependencies

Source: https://langfuse.com/docs/get-started

Install the required AI SDK, OpenTelemetry, and Langfuse integration packages.

```bash
npm install ai @ai-sdk/openai @langfuse/vercel-ai-sdk @langfuse/tracing @langfuse/otel @opentelemetry/sdk-node
```

--------------------------------

### Initialize Langfuse Client for Python

Source: https://langfuse.com/docs/prompt-management/get-started

Basic setup for using Langfuse with Langchain in Python.

```python
from langfuse import Langfuse
from langchain_core.prompts import ChatPromptTemplate

# Initialize Langfuse client
langfuse = Langfuse()
```

--------------------------------

### Initialize Langfuse and OpenAI Client

Source: https://langfuse.com/docs/prompts

Setup for using Langfuse with OpenAI in Python.

```python
import openai
from langfuse import get_client

# Initialize Langfuse client
langfuse = get_client()
```

--------------------------------

### Install OpenLLMetry and Langfuse dependencies

Source: https://langfuse.com/docs/opentelemetry/example-openllmetry

Install the required Python packages for OpenAI interaction, OpenLLMetry instrumentation, and Langfuse integration.

```bash
%pip install openai traceloop-sdk langfuse
```

--------------------------------

### Install Langfuse and Vercel AI SDK dependencies

Source: https://langfuse.com/docs/prompts

Install the required packages to enable Langfuse integration with the Vercel AI SDK.

```bash
npm install @langfuse/client ai
```

--------------------------------

### Execute the application with sample input

Source: https://langfuse.com/docs/security-and-guardrails

Run the summarization function using a sample court transcript to demonstrate the end-to-end security workflow.

```python
prompt = """
Plaintiff, Jane Doe, by and through her attorneys, files this complaint
against Defendant, Big Corporation, and alleges upon information and belief,
except for those allegations pertaining to personal knowledge, that on or about
July 15, 2023, at the Defendant's manufacturing facility located at 123 Industrial Way, Springfield, Illinois, Defendant negligently failed to maintain safe working conditions,
leading to Plaintiff suffering severe and permanent injuries. As a direct and proximate
result of Defendant's negligence, Plaintiff has endured significant physical pain, emotional distress, and financial hardship due to medical expenses and loss of income. Plaintiff seeks compensatory damages, punitive damages, and any other relief the Court deems just and proper.
"""
summarize_transcript(prompt)
```

--------------------------------

### Install RAGAS dependencies

Source: https://langfuse.com/docs/evaluation/features/synthetic-datasets

Install the necessary packages for RAGAS and Langchain integration.

```bash
%pip install ragas langchain-community langchain-openai unstructured
```

--------------------------------

### Java SDK Installation and Usage

Source: https://langfuse.com/docs/api-and-data-platform/features/public-api

Configure the Java SDK via Maven dependencies and instantiate the client with region-specific URLs and credentials.

```xml
<dependencies>
  <dependency>
    <groupId>com.langfuse</groupId>
    <artifactId>langfuse-java</artifactId>
    <version>0.0.1-SNAPSHOT</version>
  </dependency>
</dependencies>

<repositories>
  <repository>
    <id>github</id>
    <name>GitHub Package Registry</name>
    <url>https://maven.pkg.github.com/langfuse/langfuse-java</url>
  </repository>
</repositories>
```

```java
import com.langfuse.client.LangfuseClient;
import com.langfuse.client.resources.prompts.types.PromptMetaListResponse;
import com.langfuse.client.core.LangfuseClientApiException;

LangfuseClient client = LangfuseClient.builder()
  .url("https://cloud.langfuse.com") // 🇪🇺 EU data region
  // Other Langfuse data regions:
  // .url("https://us.cloud.langfuse.com") // 🇺🇸 US
  // .url("https://jp.cloud.langfuse.com") // 🇯🇵 Japan
  // .url("https://hipaa.cloud.langfuse.com") // ⚕️ HIPAA
  // .url("http://localhost:3000") // 🏠 Local deployment
  .credentials("pk-lf-...", "sk-lf-...")
  .build();

try {
  PromptMetaListResponse prompts = client.prompts().list();
} catch (LangfuseClientApiException error) {
  System.out.println(error.getBody());
  System.out.println(error.getStatusCode());
}
```

--------------------------------

### Instantiate and Use Java SDK

Source: https://langfuse.com/docs/api

Initialize the LangfuseClient with credentials and a base URL, then handle potential API exceptions.

```java
import com.langfuse.client.LangfuseClient;
import com.langfuse.client.resources.prompts.types.PromptMetaListResponse;
import com.langfuse.client.core.LangfuseClientApiException;

LangfuseClient client = LangfuseClient.builder()
  .url("https://cloud.langfuse.com") // 🇪🇺 EU data region
  // Other Langfuse data regions:
  // .url("https://us.cloud.langfuse.com") // 🇺🇸 US
  // .url("https://jp.cloud.langfuse.com") // 🇯🇵 Japan
  // .url("https://hipaa.cloud.langfuse.com") // ⚕️ HIPAA
  // .url("http://localhost:3000") // 🏠 Local deployment
  .credentials("pk-lf-...", "sk-lf-...")
  .build();

try {
  PromptMetaListResponse prompts = client.prompts().list();
} catch (LangfuseClientApiException error) {
  System.out.println(error.getBody());
  System.out.println(error.getStatusCode());
}
```

--------------------------------

### Install Langfuse Dependencies

Source: https://langfuse.com/docs/get-started

Install the core Langfuse SDK and LangChain integration package.

```bash
npm install @langfuse/core @langfuse/langchain
```

--------------------------------

### Install Langfuse SDK

Source: https://langfuse.com/docs/integrations/openai/js/get-started

Install the required Langfuse and OpenAI packages via npm.

```bash
npm install @langfuse/openai openai
```

--------------------------------

### Manual Skill Installation

Source: https://langfuse.com/docs/api-and-data-platform/features/agent-skill

Manually clone the repository and symlink the skill folder into your agent's skills directory.

```bash
git clone https://github.com/langfuse/skills.git /path/to/langfuse-skills
```

```bash
mkdir -p /path/to/<agent-skill-root>/skills
```

```bash
ln -s /path/to/langfuse-skills/skills/langfuse /path/to/<agent-skill-root>/skills/langfuse
```

--------------------------------

### Install Required Dependencies

Source: https://langfuse.com/docs/evaluation/features/synthetic-datasets

Install the necessary LangChain and Langfuse packages for the RAG pipeline.

```bash
# Install dependencies
%pip install --upgrade langchain-community langchain-openai langchain-chroma langfuse "unstructured[md]"
```

--------------------------------

### Configure Environment in Python SDK

Source: https://langfuse.com/docs/observability/features/environments

Demonstrates setting the environment via environment variables, client initialization, and request-scoped propagation using propagate_attributes.

```python
from langfuse import get_client, observe, propagate_attributes
import os

# Set the environment variable
# Alternatively, set via .env file and load via dotenv
os.environ["LANGFUSE_TRACING_ENVIRONMENT"] = "production"

# Get the client (will use environment variable)
langfuse = get_client()

# All operations will now be associated with the "production" environment
with langfuse.start_as_current_observation(as_type="span", name="my-operation") as span:
    # Your code here
    pass

@observe
def main():
    return "Hello"

main()

# For request-scoped environments, propagate the environment explicitly.
# This maps to the first-class langfuse.environment field.
with langfuse.start_as_current_observation(as_type="span", name="proxy-request"):
    with propagate_attributes(environment="staging"):
        # All child observations created here are associated with staging.
        pass
```

--------------------------------

### Install Langfuse OpenAI package

Source: https://langfuse.com/docs/get-started

Install the wrapper package to enable automatic tracing of OpenAI calls.

```bash
npm install @langfuse/openai
```

--------------------------------

### Install via Skills CLI

Source: https://langfuse.com/docs/api-and-data-platform/features/agent-skill

Use the skills CLI to add the Langfuse skill to your environment, optionally targeting a specific agent ID.

```bash
npx skills add langfuse/skills --skill "langfuse"
```

```bash
npx skills add langfuse/skills --skill "langfuse" --agent "<agent-id>"
```

--------------------------------

### Clone sample documents

Source: https://langfuse.com/docs/evaluation/features/synthetic-datasets

Download sample markdown documents for RAGAS test set generation.

```bash
!git clone https://huggingface.co/datasets/explodinggradients/Sample_Docs_Markdown
```

--------------------------------

### Create and fetch datasets in folders

Source: https://langfuse.com/docs/datasets/get-started

Use the full path string to create or retrieve datasets organized within folders.

```python
dataset_name = "evaluation/qa-dataset"

# When creating a dataset, use the full dataset name
langfuse.create_dataset(
    name=dataset_name,
)

# When fetching a dataset in a folder, use the full dataset name
langfuse.get_dataset(
    name=dataset_name
)
```

```typescript
import { LangfuseClient } from "@langfuse/client";

const langfuse = new LangfuseClient();

const datasetName = "evaluation/qa-dataset";
const encodedName = encodeURIComponent(datasetName); // "evaluation%2Fqa-dataset"

// When creating a dataset, use the full dataset name
await langfuse.dataset.create(datasetName);

// When fetching a dataset in a folder, use the encoded name
await langfuse.dataset.get(encodedName);
```

--------------------------------

### Reference prompts in SDKs/API

Source: https://langfuse.com/docs/prompt-management/features/composability

Use these formats to embed other prompts by name and version or label.

```bash
@@@langfusePrompt:name=PromptName|version=1@@@
```

```bash
@@@langfusePrompt:name=PromptName|label=production@@@
```

--------------------------------

### Agent Instruction for Skill Installation

Source: https://langfuse.com/docs/get-started

Instruction to provide to a coding agent to install the Langfuse skill and configure tracing.

```text
Install the Langfuse Agent Skill from github.com/langfuse/skills
and use it to add tracing to this application with Langfuse
following best practices.
```

--------------------------------

### Install Langfuse Docs MCP in Windsurf

Source: https://langfuse.com/docs/docs-mcp

Configure the MCP server using the npx command as a local proxy.

```json
{
  "mcpServers": {
    "langfuse-docs": {
      "command": "npx",
      "args": ["mcp-remote", "https://langfuse.com/api/mcp"]
    }
  }
}
```

--------------------------------

### Install Langfuse CLI via AI agent

Source: https://langfuse.com/docs/api-and-data-platform/features/cli

Instruct an AI coding agent to install the CLI package.

```markdown
please install langfuse-cli
```

--------------------------------

### Install Dependencies for OpenInference Tracing

Source: https://langfuse.com/docs/opentelemetry/example-arize

Install the required Python packages for OpenTelemetry tracing, OpenInference instrumentation, and the OpenAI SDK.

```bash
%pip install langfuse openai openinference-instrumentation-openai --upgrade
```

--------------------------------

### Initialize multiple Langfuse clients

Source: https://langfuse.com/docs/observability/sdk/advanced-features

Create separate client instances for different projects to enable multi-project routing.

```python
from langfuse import Langfuse

# Initialize clients for different projects
project_a_client = Langfuse(
    public_key="pk-lf-project-a-...",
    secret_key="sk-lf-project-a-...",
    base_url="https://cloud.langfuse.com"
)

project_b_client = Langfuse(
    public_key="pk-lf-project-b-...",
    secret_key="sk-lf-project-b-...",
    base_url="https://cloud.langfuse.com"
)
```
Tokens:
Raw