[metadata]
description: Trace, evaluate, and improve AI agents with one open platform. Use production data to understand behavior, collaborate on fixes, and ship better quality at lower cost and latency.
og:description: Trace, evaluate, and improve AI agents with one open platform. Use production data to understand behavior, collaborate on fixes, and ship better quality at lower cost and latency.
og:image: https://langfuse.com/api/og?title=Langfuse+%E2%80%93+Open+Source+Agent+Evals+%26+Observability&description=Trace%2C+evaluate%2C+and+improve+AI+agents+with+one+open+platform.+Use+production+data+to+understand+behavior%2C+collaborate+on+fixes%2C+and+ship+better+quality+at+lower+cost+and+latency.
og:title: Langfuse
twitter:card: summary_large_image
twitter:description: Trace, evaluate, and improve AI agents with one open platform. Use production data to understand behavior, collaborate on fixes, and ship better quality at lower cost and latency.
twitter:image: https://langfuse.com/api/og?title=Langfuse+%E2%80%93+Open+Source+Agent+Evals+%26+Observability&description=Trace%2C+evaluate%2C+and+improve+AI+agents+with+one+open+platform.+Use+production+data+to+understand+behavior%2C+collaborate+on+fixes%2C+and+ship+better+quality+at+lower+cost+and+latency.
twitter:site: langfuse.com
twitter:title: Langfuse
viewport: width=device-width, initial-scale=1

[document-links]
.NET: /integrations/native/opentelemetry
/
/discord
/users/canva
/users/khan-academy
/users/sumup
22,000+ GitHub stars: https://github.com/langfuse/langfuse
5,000+ Discord members: https://langfuse.com/discord
50M+ SDK installs/month: /careers#public-metrics
AI Engineering Library: /library
API Reference: /docs/api-and-data-platform/overview
AWS (Terraform): /self-hosting/deployment/aws
About Us: /about
Academy: /academy
Agno Agents: /integrations/frameworks/agno-agents
All product features MIT licensed: /self-hosting
Altalogy: https://altalogy.com/?ref=langfuse
Amazon AgentCore: /integrations/frameworks/amazon-agentcore
Amazon Bedrock: /integrations/model-providers/amazon-bedrock
Anthropic: /integrations/model-providers/anthropic
Async ingestion via Redis queue: /self-hosting/deployment/infrastructure/cache
AutoGen: /integrations/frameworks/autogen
Azure (Terraform): /self-hosting/deployment/azure
Azure OpenAI: /integrations/model-providers/openai-py
Blog: /blog
Canva: /users/canva
Careers: /careers
Changelog: /changelog
Claude Agent SDK (JS): /integrations/frameworks/claude-agent-sdk-js
Claude Agent SDK (Python): /integrations/frameworks/claude-agent-sdk
Claude Code: /integrations/developer-tools/claude-code
ClickHouse Agentic Data Stack: /integrations/other/agentic-data-stack
Clickhouse OLAP database: /self-hosting/deployment/infrastructure/clickhouse
Community Q&A threads 1.8k: https://github.com/orgs/langfuse/discussions/categories/support
Configure MCP →: /docs/api-and-data-platform/features/mcp-server
Configure the CLI →: /docs/api-and-data-platform/features/cli
Consistent evaluator sampling 2 days ago: /changelog/2026-08-05-deterministic-evaluator-sampling
Contributors 300+: https://github.com/langfuse/langfuse/graphs/contributors
Cookie Policy: /cookie-policy
Cost & Latency Monitor cost, latency, and quality with dashboards and automated alerts.: /docs/analytics
CrewAI: /integrations/frameworks/crewai
Cursor: /integrations/developer-tools/cursor
Customers: /users
DSPy: /integrations/frameworks/dspy
Dify: /integrations/no-code/dify
Docker Compose: /self-hosting/deployment/docker-compose
Docs: /docs
Documentation D: /docs
Documentation: /docs
EU & US Data Regions: /security/data-regions
Edge-cached prompts: /docs/prompt-management/features/caching
Enterprise: /enterprise
Evaluation LLM-as-a-judge, heuristic functions, or human review. Run evaluators on production data or during experiments.: /docs/scores
Evaluation: /docs/evaluation/overview
Evaluations: /docs/evaluation/overview
Events: /events
Example Project: /docs/demo
Experiments Define test cases and run experiments. Compare results side by side.: /docs/experimentation
Fork, modify, contribute: https://github.com/langfuse/langfuse
GCP (Terraform): /self-hosting/deployment/gcp
GDPR: /security/gdpr
Get Demo G: /talk-to-us
GitHub Stars 32.6k: https://github.com/langfuse/langfuse
Go: /integrations/native/opentelemetry
Google ADK: /integrations/frameworks/google-adk
Google Gemini: /integrations/model-providers/google-gemini
Google Vertex AI: /integrations/model-providers/google-vertex-ai
Groq: /integrations/model-providers/groq
Guides & Cookbooks: /guides
HIPAA-ready region: /security/hipaa
Handbook: /handbook
Human Annotation Collaborative Human-in the-Loop workflows to review traces and create golden datasets.: /docs/evaluation/evaluation-methods/annotation-queues
ISO 27001: /security/iso27001
Imprint: /imprint
Install the skill →: /docs/api-and-data-platform/features/agent-skill
Integrations: /integrations
Interactive Demo: /docs/demo
Java: /integrations/native/opentelemetry
Kubernetes (Helm): /self-hosting/deployment/kubernetes-helm
LLM Observability: /docs/observability/overview
LangChain DeepAgents: /integrations/frameworks/langchain-deepagents
LangChain: /integrations/frameworks/langchain
Langflow: /integrations/no-code/langflow
Langfuse CLI: /docs/api-and-data-platform/features/cli
Langfuse for Agents: /agents
Langfuse v4: up to 165× faster · Read more Langfuse v4 is here: real-time, up to 165× faster · Read more: /docs/v4
Latest OSS release today: https://github.com/langfuse/langfuse/releases
Launch App L: /cloud
Learn in Academy: /academy
LiteLLM (Proxy/Gateway): /integrations/gateways/litellm
LiteLLM: /integrations/frameworks/litellm-sdk
LiveKit: /integrations/frameworks/livekit
LlamaIndex: /integrations/frameworks/llamaindex
Mastra: /integrations/frameworks/mastra
Metrics: /docs/metrics/overview
Microsoft Agent Framework: /integrations/frameworks/microsoft-agent-framework
Mistral AI: /integrations/model-providers/mistral-sdk
Observability Hierarchical traces capture every LLM call, tool invocation, and retrieval step. Filter by user, session, cost, latency, or custom metadata.: /docs/tracing
Observability: /docs/observability/overview
Ollama: /integrations/model-providers/ollama
Open source: /handbook/chapters/open-source
OpenAI Agents SDK: /integrations/frameworks/openai-agents
OpenAI: /integrations/model-providers/openai-py
OpenClaw: /integrations/other/openclaw
OpenRouter: /integrations/gateways/openrouter
OpenWebUI: /integrations/no-code/openwebui
Overview: /docs
PHP: /integrations/native/opentelemetry
Platform MCP Server: /docs/api-and-data-platform/features/mcp-server
Playground Test prompts on real production inputs and compare models side-by-side.: /docs/playground
Playground: /docs/prompt-management/features/playground
PostHog: /integrations/analytics/posthog
Press: /press
Pricing: /pricing
Privacy: /privacy
Prompt Management Separate prompts from code with one-click deployments and rollbacks. Turn improving your production prompts a team sport.: /docs/prompts
Prompt Management: /docs/prompt-management/overview
Promptfoo: /integrations/other/promptfoo
Pulse: find the outliers in your traces 10 days ago: /changelog/2026-07-28-langfuse-pulse
Pydantic AI: /integrations/frameworks/pydantic-ai
Python (Native SDK): /docs/observability/sdk/overview
Query SDK: /docs/api-and-data-platform/features/query-via-sdk
RAGflow: /integrations/no-code/ragflow
REST APIs for everything: /docs/api-and-data-platform/features/public-api
Ragas: /integrations/frameworks/ragas
Reach out to Support: /support
Read story: /users/canva
Read story: /users/khan-academy
Read story: /users/merckgroup
Read story: /users/sumup
Request it →: /integrations#request-integration
Roadmap threads 1.6k: https://github.com/orgs/langfuse/discussions/categories/ideas
Roadmap: /docs/roadmap
Ruby: /integrations/native/opentelemetry
S3 blob storage export: /docs/api-and-data-platform/features/export-to-blob-storage
S3/Blob storage for large payloads: /self-hosting/deployment/infrastructure/blobstorage
SDKs: /docs/observability/sdk/overview
SKILL.md: /docs/api-and-data-platform/features/agent-skill
SOC 2 Type II: /security/soc2
Scales to billions of monthly events: /self-hosting/configuration/scaling
Secure remote experiment triggers 10 days ago: /changelog/2026-07-28-secure-remote-experiment-triggers
Security: /security
See all integrations: /integrations
Self-Hosting: /self-hosting
Spring AI: /integrations/frameworks/spring-ai
Start free S: /cloud
Status: https://status.langfuse.com
Strands Agents: /integrations/frameworks/strands-agents
Support: /support
Swift: /integrations/native/opentelemetry
Talk to Sales: /talk-to-us
Talk to Us: /talk-to-us
Temporal: /integrations/frameworks/temporal
Terms: /terms
TypeScript (Native SDK): /docs/observability/sdk/overview
Users: /users
Vercel AI SDK: /integrations/frameworks/vercel-ai-sdk
View All: /changelog
View documentation →: /docs/langfuse-assistant
Walkthroughs: /guides
Weekly releases and community hours: /changelog
Workshop: /workshop
analytics dashboards: /docs/metrics/overview
by ClickHouse: https://clickhouse.com
evaluations: /docs/evaluation/overview
https://github.com/langfuse/langfuse
https://www.linkedin.com/company/langfuse/
https://www.youtube.com/@langfuse
https://x.com/langfuse
langfuse ClickHouse Langfuse joins ClickHouse Our goal continues to be building the best AI engineering platform Read story: /blog/joining-clickhouse
langfuse Get Started with Tracing Step-by-step guide to ingesting your first trace using OpenAI, LangChain, or the SDKs. Get Started with Tracing This guide walks you through ingesting your first trace. Read docs: /docs/get-started
n8n: /integrations/no-code/n8n
open-source: https://github.com/langfuse/langfuse
prompt management: /docs/prompt-management/overview
public demo project: /docs/demo
sign up for free: /cloud
tracing: /docs/observability/overview
vLLM: /integrations/model-providers/vllm
xAI: /integrations/model-providers/xai-grok
🐐 Hiring in Europe and SF Looking for GOATS!: /careers

[content]
Langfuse
Langfuse v4: up to 165× faster ·
Read more
Langfuse v4 is here: real-time, up to 165× faster ·
Read more
by ClickHouse
🐐
Hiring in Europe and SF
Looking for GOATS!
Product
Overview
LLM Observability
Prompt Management
Evaluation
Metrics
langfuse
Get Started with Tracing
Step-by-step guide to ingesting your first trace using OpenAI, LangChain, or the SDKs.
Get Started with Tracing
This guide walks you through ingesting your first trace.
Read docs
Resources
Academy
Workshop
Blog
Changelog
Roadmap
Users
Example Project
Walkthroughs
Support
langfuse
ClickHouse
Langfuse joins ClickHouse
Our goal continues to be building the best AI engineering platform
Read story
Docs
Changelog
Pricing
Launch App
L
Get Demo
G
Community Stats
GitHub Stars
32.6k
Contributors
300+
Community Q&A threads
1.8k
Roadmap threads
1.6k
Latest OSS release
today
Changelog
View All
Consistent evaluator sampling
2 days ago
Pulse: find the outliers in your traces
10 days ago
Secure remote experiment triggers
10 days ago
Self Hosting Guides
Docker Compose
Kubernetes (Helm)
AWS (Terraform)
GCP (Terraform)
Azure (Terraform)
Used by
21
of Fortune 50
10+ billion
observations/month
100,000+
engineers building on Langfuse
Used by
21
of Fortune 50
10+ billion
observations/month
100,000+
engineers building on Langfuse
Used by
21
of Fortune 50
10+ billion
observations/month
100,000+
engineers building on Langfuse
Open Source
Agent Evals &
Observability
Trace, evaluate, and improve AI agents with one open platform. Use production data to understand behavior, collaborate on fixes, and ship better quality at lower cost and latency.
Start free
S
Documentation
D
Onboard with AI
Read story
Read story
Read story
Read story
Gain
deep visibility
into your traces
Launch,
observe,
improve
— repeat.
Langfuse connects tracing, monitoring, datasets, experiments, and evaluation in one continuous loop. Use production signals to understand behavior, test improvements, and ship better agents with confidence.
The full LLM engineering loop
See how observability, prompts, evals, experiments, and human feedback work together.
Learn in Academy
All the tools,
one
integrated platform.
One integrated platform to trace, manage prompts, evaluate, and experiment from prototype to production scale.
Observability
Hierarchical traces capture every LLM call, tool invocation, and retrieval step. Filter by user, session, cost, latency, or custom metadata.
Evaluation
LLM-as-a-judge, heuristic functions, or human review. Run evaluators on production data or during experiments.
Prompt Management
Separate prompts from code with one-click deployments and rollbacks. Turn improving your production prompts a team sport.
Playground
Test prompts on real production inputs and compare models side-by-side.
Experiments
Define test cases and run experiments. Compare results side by side.
Human Annotation
Collaborative Human-in the-Loop workflows to review traces and create golden datasets.
Cost & Latency
Monitor cost, latency, and quality with dashboards and automated alerts.
Start free
S
Canva
Canva's AI team relies on Langfuse to trace and debug their generative design features in production.
Works with
any stack.
Langfuse works with any language and framework supporting OTel instrumentation. Additionally, 100+ integrations make getting started even easier. No framework lock-in.
Languages (via OTel)
Python (Native SDK)
TypeScript (Native SDK)
Go
Java
.NET
Ruby
PHP
Swift
Agent frameworks
LangChain
Vercel AI SDK
LiteLLM
Pydantic AI
Google ADK
CrewAI
LiveKit
and many more…
Model providers
OpenAI
Anthropic
Amazon Bedrock
Azure OpenAI
Mistral AI
Google Gemini
xAI
vLLM
Groq
and many more…
100+ more integrations
Claude Code
LiteLLM (Proxy/Gateway)
OpenClaw
Claude Agent SDK (Python)
LangChain DeepAgents
OpenWebUI
Ollama
OpenAI Agents SDK
Dify
Langflow
OpenRouter
n8n
Spring AI
Cursor
PostHog
Claude Code
LiteLLM (Proxy/Gateway)
OpenClaw
Claude Agent SDK (Python)
LangChain DeepAgents
OpenWebUI
Ollama
OpenAI Agents SDK
Dify
Langflow
OpenRouter
n8n
Spring AI
Cursor
PostHog
DSPy
Amazon AgentCore
Strands Agents
LlamaIndex
Agno Agents
Temporal
ClickHouse Agentic Data Stack
Mastra
Claude Agent SDK (JS)
Promptfoo
Microsoft Agent Framework
Google Vertex AI
Ragas
AutoGen
RAGflow
DSPy
Amazon AgentCore
Strands Agents
LlamaIndex
Agno Agents
Temporal
ClickHouse Agentic Data Stack
Mastra
Claude Agent SDK (JS)
Promptfoo
Microsoft Agent Framework
Google Vertex AI
Ragas
AutoGen
RAGflow
See all integrations
Don't find your integration?
Request it →
Open platform.
Open source.
We are huge fans of open standards and data portability. Langfuse won't lock in your data, ever.
Self-host at scale
Docker Compose
Kubernetes (Helm)
AWS (Terraform)
GCP (Terraform)
Azure (Terraform)
MIT license
All product features MIT licensed
Scales to billions of monthly events
Fork, modify, contribute
APIs & exports
REST APIs for everything
Query SDK
S3 blob storage export
Active OSS community
22,000+ GitHub stars
5,000+ Discord members
Weekly releases and community hours
Made for
developers
,
loved by
agents
.
Work in the app or from your IDE. The Assistant investigates production and takes approved actions; SKILL.md, CLI, and MCP connect coding agents to Langfuse.
00
In-app
Langfuse Assistant
Automate the AI engineering loop: investigate production data, understand what happened, and turn findings into approved actions without leaving Langfuse.
Debug traces
Find failed generations and traces with high latency
Optimize spend
Break down token spend, cost, and latency
Build evals
Create regression datasets and score configs
View documentation
→
0
1
Coding agents
SKILL.md
A ready-made skill for managing prompts, traces, and evals through natural language.
Install the skill
→
0
2
Terminal
Langfuse CLI
Full API access from the terminal for agent workflows, scripts, and CI/CD.
Configure the CLI
→
0
3
IDE agents
Platform MCP Server
Structured access for IDE agents to manage prompts, query traces, and use Langfuse data.
Configure MCP
→
Enterprise
scale
and security
.
Traditional observability handles many small spans. LLM systems run differently. Every step carries rich, verbose I/O that legacy platforms can't handle at scale. Langfuse ingests and queries LLM traces reliably at enterprise scale while following strict compliance frameworks.
Architecture
Clickhouse OLAP database
Async ingestion via Redis queue
S3/Blob storage for large payloads
Edge-cached prompts
Reliability at scale
50M+ SDK installs/month
10+ billion observations processed per month
2300+ customers
99.9% uptime
Security & compliance
SOC 2 Type II
ISO 27001
GDPR
EU & US Data Regions
HIPAA-ready region
Start free
S
Canva
Canva's AI team relies on Langfuse to trace and debug their generative design features in production.
Why use
Langfuse?
Langfuse is the most widely adopted open-source LLM engineering platform. Developers who value open-source and control over their data build production grade agents and LLM applications with Langfuse.
The full cycle
Langfuse powers the entire development cycle from prototype to full scale production loads.
Unified platform
Open source (MIT)
OTel native
100+ integrations
Built for scale
Async by default
Loved by agents
Production-proven
Shipping velocity
The full cycle
Langfuse powers the entire development cycle from prototype to full scale production loads.
Unified platform
All components of Langfuse work great standalone but excel when used together.
Open source (MIT)
Inspect the code. Self-host for free. We are the largest OSS community in our category.
OTel native
Standard trace format. Works with existing OpenTelemetry instrumentation.
100+ integrations
Works with any model, any framework, and stack.
Built for scale
ClickHouse backend allows to query millions of traces in milliseconds.
Async by default
Tracing never blocks your application. Background processing, automatic batching.
Loved by agents
CLI, MCP, accessible docs - coding agents love working with Langfuse.
Production-proven
Billions of events processed per month. 50M+ SDK installs/month. Fortune 50 deployments.
Shipping velocity
The AI space is changing fast. We understand what patterns matter and ship daily.
Get Started
—
Free tier:
50k observations/month
. No credit card required.
Start improving
your agents
in under 5 minutes.
Get Started
—
Free tier:
50k observations/month
. No credit card required.
Start free
S
Documentation
D
Install via Coding Agent
Manual Install
Claude Code
Cursor
Codex
or any other agent
Tracing:
Prompt:
Install the Langfuse Agent Skill from github.com/langfuse/skills and use it to add tracing to this application with Langfuse following best practices.
Evals:
Prompt:
Install the Langfuse Agent Skill from github.com/langfuse/skills and use it to set up evals for this application with Langfuse. Guide me through choosing the right evaluation approach methods.
Prompt Management:
Prompt:
Install the Langfuse Agent Skill from github.com/langfuse/skills and use it to migrate the prompts in this codebase to Langfuse.
Need help? —
Talk to Sales
·
Reach out to Support
Questions & Answers
What is Langfuse?
Langfuse is an
open-source
AI engineering platform that helps teams build, monitor, and improve their LLM applications. It covers the full development lifecycle with
tracing
,
prompt management
,
evaluations
, and
analytics dashboards
— all in one place. Langfuse is used by 2,300+ companies and processes billions of observations per month. You can try it instantly with the
public demo project
or
sign up for free
What does Langfuse help me with?
Can I use just tracing without the other features?
What deployment options do exist?
Is self-hosting actually free?
What frameworks are supported?
What's the latency impact?
Is Langfuse secure and compliant?
How do I get started?
How does pricing work?
Product
Observability
Prompt Management
Evaluations
Metrics
Langfuse for Agents
Playground
Pricing
Enterprise
Developers
Documentation
Self-Hosting
SDKs
Integrations
API Reference
Status
Talk to Us
Resources
Blog
Changelog
Events
Roadmap
Interactive Demo
Customers
AI Engineering Library
Workshop
Guides & Cookbooks
Company
About Us
Careers
Handbook
Press
Security
Support
Open source
Terms
Privacy
Imprint
Cookie Policy
© 2022–
2026
Langfuse GmbH
/ Finto Technologies Inc.
Design by
Altalogy
Ask AI
A
A
Auto-advance is
active
. Press Escape to
pause
auto-advance.
Gain deep visibility into your traces
Track model cost and latency
Improve your prompts
Evaluate model outputs automatically
Collaborate on human reviews
Iterate with structured experiments
Gain deep visibility into your traces
