Langfuse
Used by 19 of Fortune 50
10+ billion observations/month
100,000+ engineers building on Langfuse
Used by 19 of Fortune 50
10+ billion observations/month
100,000+ engineers building on Langfuse
Used by 19 of Fortune 50
10+ billion observations/month
100,000+ engineers building on Langfuse
Open SourceAIEngineeringPlatform
Debug LLM applications and agents in minutes. Spot issues before your users do. Collaborate with your team to continuously improve on cost, latency and quality. Any model, any framework. Based on OpenTelemetry.
Gain deep visibility into your traces
Launch, observe, improve — repeat.
Langfuse helps you ship AI Agents/Products from prototype to production and beyond. Once in production we power your continous improvement loop using production data to make your agents and LLM applications ever more powerful.
The full LLM engineering loop
Langfuse brings observability, prompts, evals, experiments, and human annotation into one connected workflow — so you can move from prototype to production and keep improving with real usage data. Hover any part of the diagram to learn more.
All the tools, oneintegrated platform.
One integrated platform to trace, manage prompts, evaluate, and experiment from prototype to production scale.
ObservabilityHierarchical traces capture every LLM call, tool invocation, and retrieval step. Filter by user, session, cost, latency, or custom metadata.
EvaluationLLM-as-a-judge, heuristic functions, or human review. Run evaluators on production data or during experiments.
Prompt ManagementSeparate prompts from code with one-click deployments and rollbacks. Turn improving your production prompts a team sport.
PlaygroundTest prompts on real production inputs and compare models side-by-side.
ExperimentsDefine test cases and run experiments. Compare results side by side.
Human AnnotationCollaborative Human-in the-Loop workflows to review traces and create golden datasets.
Cost & LatencyMonitor cost, latency, and quality with dashboards and automated alerts.
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, 80+ integrations make getting started even easier. No framework lock-in.
Languages (via OTel)
PythonPython (Native SDK)TypescriptTypeScript (Native SDK)GoJava.NETRubyPHPSwift
Agent frameworks
LangChain
Vercel AI SDK
LiteLLM
Pydantic AI
Google ADK
CrewAI
LiveKitand many more…
Model providers
OpenAI
Anthropic
Amazon Bedrock
Azure OpenAI
Mistral AI
Google Gemini
xAI
vLLM
Groqand many more…
80+ more integrations
Claude CodeLiteLLM (Proxy/Gateway)
OpenClaw
Claude Agent SDK (Python)LangChain DeepAgents
OpenWebUI
OllamaOpenAI Agents SDK
Dify
Langflow
OpenRouter
n8n
Spring AI
Cursor
PostHogClaude CodeLiteLLM (Proxy/Gateway)OpenClawClaude Agent SDK (Python)LangChain DeepAgentsOpenWebUIOllamaOpenAI Agents SDKDifyLangflowOpenRoutern8nSpring AICursorPostHog
DSPyAmazon AgentCore
Strands Agents
LlamaIndex
Agno Agents
Temporal
Agentic Data Stack
MastraClaude Agent SDK (JS)
PromptfooMicrosoft Agent Framework
Google Vertex AI
Ragas
AutoGen
RAGflowDSPyAmazon AgentCoreStrands AgentsLlamaIndexAgno AgentsTemporalAgentic Data StackMastraClaude Agent SDK (JS)PromptfooMicrosoft Agent FrameworkGoogle Vertex AIRagasAutoGenRAGflow
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
APIs & exports
Active OSS community
Made for developers, loved by agents.
Langfuse works by default with your coding agents. Install our MCP servers and CLI to develop at the speed of thought. Let Claude Code, Cursor and Codex do the hard work.
A ready-made skill for your coding agent. Manage prompts, traces, and evals through natural language — no manual API calls needed.
Full API access from the terminal. Let coding agents manage Langfuse for you, or script your workflows in CI/CD.
Interact with your Langfuse data programmatically from your IDE. Manage prompts, query traces, and more.
Enterprise scaleand 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
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
80+ 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.
- 80+ 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 improvingyour agents
in under 5 minutes.
Get Started — Free tier: 50k observations/month. No credit card required.
Claude Code
CursorCodex
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