Langfuse

Langfuse is an open-source AI engineering platform that traces, evaluates, and improves LLM applications.

Langfuse connects tracing, monitoring, datasets, experiments, and evaluation in one continuous loop to help teams build and improve AI agents. It features hierarchical traces for LLM calls and tool invocations, prompt management with one-click deployments, a playground to test prompts, experiments, and human annotation workflows. The platform works with any language and framework supporting OpenTelemetry instrumentation alongside over 100 integrations. Built for scale, it uses a ClickHouse OLAP database and asynchronous ingestion via a Redis queue. It is designed for developers building production-grade LLM applications who value data portability and open-source standards.

Langfuse at a glance

Best for: Software engineer
Runs on: Web, API, CLI, self-hosted, open source
Pricing: Hobby plan free with 50k units; Core plan; Pro plan; Enterprise plan with custom pricing
Released: May 2023
Category: AI coding tool
Website: langfuse.com

Key features

Hierarchical traces for LLM calls and tool invocations
Prompt management with one-click deployments and rollbacks
Playground to test prompts on production inputs
Experiments to compare results side by side
Collaborative human-in-the-loop annotation workflows
Cost and latency monitoring with dashboards
REST APIs and query SDK for data access

Use it for

Trace and debug generative design features
Manage and test production prompts
Run evaluations using LLM-as-a-judge or human review
Monitor token spend, cost, and latency

Companion to

Tools whose articles name Langfuse as a companion, each with its job for them.

CrewAI — Langfuse: Self-hosted tracing
LangChain — Langfuse: Open-source tracing
n8n — Langfuse: Tracing

Alternatives to Langfuse

Langfa.st — Compare model responses side-by-side to validate prompts before deployment.
LangChain4j — Integrates LLMs and vector stores directly into Java applications.
LiteLLM — One OpenAI-format API for 100+ LLM providers
LangSmith — Step-by-step traces and evals that show what an agent actually did
LangChain — Compose exactly the agent your use case needs from model, tools, prompt, and middleware.
LangWatch — Runs simulations and continuous evaluations to make AI agents reliable.
llm-d — Deploys production-grade LLM serving on existing hardware clusters.
WebLLM — Runs high-performance language models directly inside web browsers.
Langroid — Simplifies multi-agent orchestration and app development with intuitive abstractions.
OptiLLM — Increases LLM reasoning accuracy by up to 10x without model training.
OpenLIT — It enables comprehensive monitoring, evaluation, and management of AI agents via OpenTelemetry.
llmware — Build private, local RAG pipelines with 300+ prepackaged models and integrated document parsing.

All Langfuse alternatives