Self-hosted (open source)
Free
- MIT-licensed core, deployed with Docker Compose or Kubernetes
What Is Langfuse? Pricing, Features, Pros & Cons (2026)
Founded 2022|Berlin, Germany|Free open-source self-hosting; cloud plans from free to $2,499/month, metered by units
What is Langfuse?
Langfuse is an open-source LLM engineering platform for tracing, evaluation, prompt management, and metrics, with an MIT-licensed core you can self-host or use as a managed cloud. It was founded in Berlin in 2022, and ClickHouse acquired Langfuse in January 2026, with its open-source and self-hosting commitments kept in place.
Tracing covers LLM calls, tool runs, and agent steps through native Python and TypeScript SDKs and OpenTelemetry, with integrations for frameworks like LangChain, LlamaIndex, and the Vercel AI SDK. On top of those traces, Langfuse runs LLM-as-a-judge and custom evaluations, datasets and experiments, annotation queues for human review, and versioned prompts with a playground.
Cloud pricing meters units, with a free Hobby plan at 50k units a month and paid plans from $29 that include 100k units and bill $8 per additional 100k. Langfuse observes requests asynchronously rather than acting as an LLM proxy, so routing, failover, and spend limits come from a separate gateway, which is a common reason teams weigh Langfuse alternatives.
Key features
Core capabilities this platform advertises.
- Tracing for LLM calls, tool runs, and agent steps
- Python and TypeScript SDKs plus OpenTelemetry
- LLM-as-a-judge and custom evaluations
- Datasets and experiments
- Prompt management and playground
Strengths and tradeoffs
What this tool does well, and the limitations to keep in mind.
Pros
- MIT-licensed core, self-hostable with Docker Compose or Kubernetes
- Tracing, evals, and prompt management in one tool
- Unlimited users on paid cloud plans
- Free Hobby plan with 50k units a month
- Native SDKs plus OpenTelemetry for other languages
Cons
- No LLM gateway, so routing, failover, and spend limits need a separate tool
- Hobby plan keeps 30 days of data and caps at 2 users
- Enterprise SSO and RBAC need the $300/month Teams add-on on Pro
- Self-hosting means running and scaling Langfuse's services yourself
Plans & pricing
What's included in each plan, and how the tiers compare.
Hobby
Free
- 50k units/month, 30-day data access, 2 users, community support
Core
$29
Per month
- 100k units/month then $8 per 100k, 90-day data access, unlimited users, in-app support
Pro
$199
Per month
- 100k units/month then $8 per 100k, 3-year data access, data retention management, SOC 2 and ISO 27001 reports, optional $300/month Teams add-on
Enterprise
$2,499
Per month
- Audit logs, SCIM API, custom rate limits, uptime SLA, dedicated support engineer
Common use cases
Teams who want open-source LLM observability they can self-host and customize
- Tracing and debugging LLM apps and agents
- Self-hosting LLM observability
- Versioning prompts across environments
- Scoring outputs with LLM judges and human review
Using Langfuse with Respan
Respan has a native Langfuse integration. The respan-instrumentation-langfuse package patches Langfuse's OTLP exporter, so traces from your existing @observe decorators and langfuse.trace() calls go to Respan with no changes to your instrumentation.
- Keep your existing Langfuse decorators and trace calls
- Evals on live production traffic, linked to each trace
- Monitors that alert in Slack, email, or a webhook when a metric breaches
- One API for 1,000+ models with automatic fallbacks
- Hard spend limits per key, per customer, or org-wide
- Prompt versions that ship without a code deploy
Route, observe, and evaluate every LLM call
10KFree traces/mo
1,000+Models
5 minSetup
Respan routes every call and scores it on live traffic, so the trace, cost, and eval score sit in one place. Your existing Langfuse decorators send traces to Respan with no code changes.
Try Respan freeBest Langfuse alternatives & competitors
Top companies in Observability, Prompts & Evals you can use instead of Langfuse.
Respan
Nested traces for every LLM call, tool run, and agent turn
LangSmith
Native tracing for LangChain and LangGraph
Weights & Biases
ML experiment tracking
MLflow
OpenTelemetry-based tracing for LLMs and agents
Arize AI
ML observability with LLM support
Traceloop
OpenTelemetry
Datadog LLM
LLM monitoring within Datadog platform
Helicone
HoneyHive
Prompt management
Braintrust
Experiments comparing prompts and models
Phoenix
OpenTelemetry-based LLM and agent tracing
Patronus AI
Automated LLM evaluation platform
Promptfoo
Portkey
Humanloop
Ragas
RAG-specific evaluation framework
Sentry
DeepEval
LangWatch
Multi-turn agent simulation testing
Galileo AI
LLM output quality evaluation
PromptLayer
Confident AI
DeepEval open-source evaluation framework
Maxim AI
Distributed tracing for LLM and agent apps
Opik
Agenta
Future AGI
Multimodal evaluation (text, image, audio, video)
Lunary
Parea AI
Ashr
Multi-modal synthetic testing
Sentrial
Agent failure root cause analysis
Chamber
ML infrastructure automation
Moda
Hallucination detection
Athina AI
Laminar
Agent tracing with a readable transcript view
OpenObserve
Logs, metrics, traces, RUM, and session replay
AgentOps
Session replay with time-travel debugging
Fiddler AI
Agentic observability across multi-step agent workflows
Compare Langfuse
Side-by-side comparisons with other tools in this category.
Best integrations for Langfuse
Companies from adjacent layers in the AI stack that work well with Langfuse.
