Menu

Home
GitHub(在新标签页打开)

Docs

Overview
Example Project
Ask AI

Docs › Get Started

Start Tracing
Use Prompt Management
Set up Evals

Docs › Platform

Halo
Security & Guardrails
Versions & Compatibility

Docs › More

Glossary
Roadmap
Langfuse v4 is live
Docs MCP Server

Docs › Observability

Overview
Get Started
Concepts
Best Practices
Troubleshooting & FAQ

Docs › Observability › Features › Essential

Sessions
User Tracking
Environments
Tags
Metadata
Trace IDs & Distributed Tracing

Docs › Observability › Features › Advanced

Alerts
Agent Access
Comments
Corrections
User Feedback
Log Levels
Agent Graphs
Filter Search Bar
Chart any table
Pulse
Full-Text Search
Masking
MCP Tracing
Multi-Modality
Observation Types
Event queuing/batching
Releases & Versioning
Sampling
Token & Cost Tracking
Trace URLs
Web Callouts
Query Data ↗
Metrics API ↗
Custom Dashboards ↗

Docs › Observability › SDKs

Overview
Instrumentation
Advanced Features
Troubleshooting & FAQ
Upgrade Path

Docs › Observability › SDKs › Upgrade Path

Upgrade path
Python v4 → v5
Python v3 → v4
Python v2 → v3
JS/TS v5 → v6
JS/TS v4 → v5
JS/TS v3 → v4

Docs › Prompt Management

Overview
Get Started
Concepts
Troubleshooting and FAQ

Docs › Prompt Management › Features › Essential

Link to Traces
Version Control
Playground

Docs › Prompt Management › Features › Advanced

Agent Access
Variables
Prompt Composability
Message Placeholders
Config
Caching
Webhooks
GitHub Integration
n8n Node
Guaranteed Availability
A/B Testing
Folders

Docs › Evaluation

Overview
Concepts

Docs › Evaluation › Get Started

Evaluate Production Traffic
Evaluate with Datasets

Docs › Evaluation › Scores

Overview
Analytics
Data Model

Docs › Evaluation › Evaluation Methods

LLM-as-a-Judge
Decision model evaluators
Code evaluators
Annotation Queues
Scores via UI
Scores via API/SDK

Docs › Evaluation › Experiments

Data Model
Datasets
Experiments via SDK
Experiments via UI
Experiments in CI/CD
Experiments via OpenTelemetry
Compare Experiments

Docs › Evaluation › More

Agent Access
Troubleshooting and FAQ

Docs › Metrics

Overview

Docs › Metrics › Features

Custom Dashboards
Metrics API

Docs › API & Data Platform

Overview

Docs › API & Data Platform › Features

Agent Skill
CLI
MCP Server
Export from UI
Export to Blob Storage
Public API
Query via SDKs

Docs › Administration › Auth

Authentication & SSO
Access Control (RBAC)
SCIM and Org API

Docs › Administration › Security

Audit Logs
Data Deletion
Data Retention

Docs › Administration › Configuration

LLM Connections
Spend Alerts

Docs › Administration › Misc

Billable Units
Troubleshooting and FAQ

Integrations

Overview

Integrations › Native

MCP Server ↗
CLI ↗
API ↗
OpenTelemetry

Integrations › Native › OpenTelemetry

OpenTelemetry
Migrate ingestion to v4
Experiments

Integrations › Frameworks

Agno
AI SDK C++
Amazon Bedrock AgentCore
AutoGen
BeeAI
Claude Agent SDK
Claude Agent SDK (JS/TS)
CrewAI
DSPy
Eve
Flue
Google ADK
Haystack
Inferable
Instructor
Koog
LangChain DeepAgents
LangChain & LangGraph
LangGraph
LiteLLM SDK
LiveKit
LlamaIndex Workflows
LlamaIndex
Mastra
mcp-use
Microsoft Agent Framework
Mirascope
OpenAI Agents
Pipecat
Pydantic AI
Quarkus LangChain4j
Ragas
Restate
Semantic Kernel
SmolAgents
Spring AI
Strands
Swiftide
TanStack AI
Temporal
Vercel AI SDK
VoltAgent
Watsonx Orchestrate ADK

Integrations › Model Providers

Amazon Bedrock Agents
Amazon Bedrock
Anthropic (JS/TS)
Anthropic (Python)
Baseten
BytePlus
Cerebras
Claude Platform
Cleanlab
Cohere
CometAPI
Databricks
DeepSeek
Fireworks AI
Google Gemini
Google Vertex AI
Groq
Hugging Face
Mistral
Morph
Novita AI
Ollama
OpenAI Assistants API
OpenAI (JS/TS)
OpenAI (Python)
Qwen
Together AI
TypeSafe (Jev)
TypeSafe (Jev JS/TS)
vLLM
xAI Grok
Z.ai

Integrations › Gateways

Anannas
Azure API Management
Helicone
Kong Gateway
LiteLLM Proxy
OpenRouter
Portkey
Truefoundry
Vercel AI Gateway

Integrations › No-Code Agent Builders

Dify
ElevenLabs
Flowise
Codename Goose
Hermes Agent
Langdock
Langflow
LobeChat
n8n
OpenClaw
OpenWebUI
Ragflow
Vapi
Zapier

Integrations › Analytics

Coval
Mixpanel
PostHog
Trubrics

Integrations › ClickHouse Products

ClickStack
ClickHouse Data Warehouse
LibreChat
Agentic Data Stack

Integrations › Developer Tools

Augment Code
Claude Code
Codex
Cursor
GitHub Copilot
Kiro
Kiro CLI
OpenCode
Pi Agent
VS Code

Integrations › Other

Cognee
EverOS
Exa
Firecrawl
Gradio
Milvus
MLflow
OpenInference (Arize)
OpenLIT
OpenLLMetry
Parallel
Promptfoo
Slack
Tavily
Testable Minds
Weco

Self Hosting

Overview
License Key (EE)
Troubleshooting & FAQ
Configuration
Administration
Upgrade
Legacy Docs (v2)

Self Hosting › Configuration › Essential

Environment Variables
Sizing & Scaling
Hardening
FIPS mode
Backups

Self Hosting › Configuration › Advanced

Caching
Code evaluators
Custom Base Path
Encryption
Health and Readiness Check
Halo
Observability
Transactional Emails

Self Hosting › Deployment

Local/VM (Docker Compose)
Kubernetes (Helm)
AWS (Terraform)
Azure (Terraform)
GCP (Terraform)
Render
Railway

Self Hosting › Deployment › Infrastructure

Architecture Overview ↗
Application Containers
Clickhouse
Redis / Valkey
Blob Storage (S3)
PostgreSQL
LLM API / Gateway

Self Hosting › Security

Authentication and SSO
Deployment Strategies
Networking
Hardening ↗
Data Masking
Telemetry
RBAC (main docs)
Data Retention (main docs)

Self Hosting › Administration

Overview
Automated Access Provisioning
Headless Initialization
Organization Creators (EE)
UI Customization (EE)
Management API (EE)

Self Hosting › Upgrade

How to Upgrade
Versions & Compatibility
Background Migrations

Self Hosting › Upgrade › Upgrade Guides

Migrate Langfuse v3 to v4 (self-hosted)
Migrate Langfuse v2 to v3 (self-hosted)
Migrate Langfuse v1 to v2 (self-hosted)

Self Hosting › Legacy Docs (v2)

Overview
Deployment Guide
Local Deployment (docker compose)

Guides

Guides
Cookbook
Videos
ClickStack logs

Guides › Cookbook

Overview
Error Analysis
Data Migration Script
Evaluating Multi-Turn Conversations
Example - Tracing and Evaluation for the OpenAI-Agents SDK
External Evaluation Pipelines
Guide - Building an intent classification pipeline
Example - Trace and Evaluate LangGraph Agents
Example: Monitoring LLM Security
Metrics API v2
Example: Multi-modality and attachments
Agent Evaluation - How to Evaluate LLM Agents
Evaluating Multi-Turn Conversations (Simulation)
Evaluate and Improve Structured Extraction
Synthetic Datasets
JS/TS SDK Example
Prompt Management with Langchain (JS)
Example - Langfuse Prompt Management with Langchain (Python)
Example: Langfuse Prompt Management for OpenAI functions (Python)

Guides › Videos

Overview
Evals for High-Stakes AI Agents
Continuously Improving Agents With Langfuse
Headless Langfuse From Your Coding Agent
Replace Vibes With Evals: Build an Evaluation Pipeline
Debug a Fan-Out RAG Agent With Tracing
Beginner's Guide to RAG Evaluation with Langfuse and Ragas

Langfuse 简体中文

Langfuse

加宽正文