Back to vendors
L

Langfuse

Visit site
Entry priceFree Hobby (50K units/mo) · Core $29/mo · open source self hostFull pricing detail

Open-source AI engineering platform for tracing, observability, prompt management, and evaluation, self-hostable or SaaS, with an OSS community, free plan, and model-agnostic coverage.

Langfuse is an open-source AI engineering platform for tracing, evaluating and improving LLM applications and agents. Hierarchical traces capture every LLM call, tool invocation and retrieval step as trace trees and agent graphs, with session and user tracking, cost and latency, dashboards and alerts. On top of tracing it adds evaluation (datasets, experiments compared against baselines, LLM-as-a-judge and code evaluators run online or offline, human annotation queues and CI checks that block regressions) and prompt management with versioning, protected deployment labels and a playground that compares models on the customer's own LLM connections.

Langfuse works with any stack through native Python and TypeScript SDKs, OpenTelemetry and more than 100 integrations, and exposes a public API, a CLI and a platform MCP server for IDE agents. Langfuse Cloud runs in US, EU or JP data regions with a HIPAA-ready region and AWS PrivateLink, and the MIT-licensed core self-hosts on Docker, Kubernetes or Terraform. It holds SOC 2 Type II and ISO 27001, with RBAC, Enterprise SSO, SCIM and audit logs on higher plans. Langfuse has joined ClickHouse and continues to sell under its own name, from a free Hobby plan through Core at $29 a month, Pro at $199 and Enterprise at $2,499.

Vendor details

Canonical URL

https://langfuse.com

Category

Agent infrastructure

Company status

acquired

Use cases & customers

Target customers

AI developersplatform teams

Deployment options

SaaSself-hosted

In practice

Your agent works in demos but fails unpredictably in production and you can't see why. Langfuse traces every prompt, response, tool call, cost, and latency, so you can reconstruct exactly what the agent did and where it went wrong.

You need to run LLM observability on your own infrastructure for data residency reasons. Langfuse is open source and can be self hosted at production scale, with broad SDK and OpenTelemetry support so it fits your stack instead of locking you in.

You can't tell whether a prompt change actually improved quality. Langfuse lets you build datasets from real traffic, score outputs with model-based or human evaluation, and track quality over time to catch regressions before users do.

Agentic Index coverage score

8.5 / 14 capabilities · 61%

Integrations & Tool Calling Partial

More than 100 integrations with agent frameworks, model providers and gateways send traces into Langfuse, data exports to PostHog, Mixpanel and blob storage, and prompt changes send notices through webhooks and Slack. These move telemetry in and data out, and there are no connectors that let an agent take authenticated actions in outside systems.

SourceLangfuse, langfuse.com homepage and pricingread 2026-09-21

Workflow Orchestration Not documented

For agents that run elsewhere, Langfuse provides tracing, evaluation and prompt management, and prompt composability links prompts together. There are no workflows that sequence, branch or retry an agent's steps, or that mix deterministic nodes with agent steps.

SourceLangfuse, langfuse.com homepage and pricingread 2026-09-21

Knowledge Grounding & RAG Not documented

Traces, prompts and evaluation datasets are stored, but there is no retrieval structure over the customer's documents or knowledge that grounds an agent's answers. Langfuse evaluates the grounding of RAG systems built elsewhere instead of providing one.

SourceLangfuse, langfuse.com pricing and llms.txtread 2026-09-21

Human Oversight & Guardrails Partial

Protected deployment labels restrict who can promote a prompt version to production, and CI checks can require an approved baseline before a change ships, so changes to an agent's behavior pass a release gate that people control. There is no approval step, escalation rule or pause before an agent acts at run time.

SourceLangfuse, langfuse.com pricing and llms.txtread 2026-09-21

Security, Identity & Governance Full

Reports for the SOC 2 Type II and ISO 27001 attestations are available from the Pro plan, and there is a region ready for HIPAA and a GDPR DPA. Access control covers RBAC at organization and project level, sign in with Google, Azure AD or GitHub, Enterprise SSO with Okta or Entra ID and SSO enforcement on the Teams add-on, SCIM provisioning, audit logs, data masking on the client side and data retention management.

SourceLangfuse, langfuse.com pricing and homepageread 2026-09-21

Observability & Auditability Full

Hierarchical traces capture every LLM call, tool invocation and retrieval step of an agent run, so each run's steps and tool calls are visible. They appear as trace trees and agent graphs with session and user tracking, token cost and latency, and can be filtered by user, session, cost, latency or metadata. Audit logs are kept separately on Enterprise, data can be exported in batches or on a schedule to blob storage and to PostHog or Mixpanel, and historical data access runs 30 days on Hobby, 90 days on Core and 3 years on Pro.

SourceLangfuse, langfuse.com homepage and pricingread 2026-09-21

Memory & State Persistence Not documented

Sessions, users and traces are recorded for inspection, and there is no session, conversation or long term memory that an agent reads back as context.

SourceLangfuse, langfuse.com pricing and homepageread 2026-09-21

Deployment & Data Residency Full

Customers of Langfuse Cloud choose US, EU or JP data regions, and there is a region ready for HIPAA and AWS PrivateLink. The MIT licensed platform self hosts with Docker Compose, Kubernetes via Helm, or Terraform on AWS, GCP and Azure, with paid Enterprise additions for self hosted deployments.

SourceLangfuse, langfuse.com pricing, homepage and llms.txtread 2026-09-21

Prebuilt Agents, Templates & Packs Partial

The Langfuse Assistant in the app is the one prebuilt agent, and a Langfuse skill packages best practice workflows for instrumentation, prompt management and API access into a customer's coding agent. Beyond those two, there is no set of prebuilt agents or templates for a customer to choose from.

SourceLangfuse, langfuse.com llms.txt and pricingread 2026-09-21

Triggers & Channel Coverage Full

Online evaluation runs LLM as a judge and code evaluators on production observations as they arrive, and monitors and alerts fire on metric and evaluator thresholds. Evaluation work therefore starts from incoming telemetry with no person starting each run, and prompt changes also emit webhooks.

SourceLangfuse, langfuse.com homepage and pricingread 2026-09-21

Model Flexibility & Routing Full

Customers configure their own LLM connections, which the Playground uses to test prompts on real production inputs and compare models side by side, and which power the LLM-as-a-judge evaluators the customer sets up. The customer brings its own providers and chooses the models Langfuse's own LLM features run on.

SourceLangfuse, langfuse.com homepage, pricing and llms.txt (LLM Connections)read 2026-09-21

APIs, SDKs & MCP Extensibility Full

An extensive public API has published rate limits by plan, alongside native Python and TypeScript SDKs, OpenTelemetry ingestion for Java, Go and other languages, a CLI, and a platform MCP server that lets IDE agents manage prompts and query traces. Stable APIs based on IDs manage evaluators, and there is an agent skill.

SourceLangfuse, langfuse.com homepage, pricing and llms.txtread 2026-09-21

Testing, Debugging & Optimization Full

Evaluation of the customer's application runs online and offline. It covers datasets, experiments from the SDK or UI compared side by side against baselines, LLM as a judge and code evaluators managed by API, custom scores, user feedback and human annotation queues, alerts on evaluator results, and CI checks that block regressions against thresholds and approved baselines. The customer's agent is tested with datasets before production, and output quality is scored over time.

SourceLangfuse, langfuse.com llms.txt, homepage and pricingread 2026-09-21

Browser & Computer Use Not documented

Langfuse observes and evaluates agents, and there is no browser, desktop or computer control by an agent.

SourceLangfuse, langfuse.com homepageread 2026-09-21

The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded

Product updates

·Deployment / data residencyVerified

FIPS mode is now available in self hosted Government deployments of Langfuse from v4.71.0. It runs Langfuse's cryptography, including TLS to its databases, hashing and application encryption, on FIPS 140-3 validated modules, and the app refuses to start if the host is not in FIPS mode. From the same version, the web and worker images for every deployment are built on Iron Bank's hardened Red Hat UBI 9.

Bears on: Security / enterprise

View source
·Security / enterpriseVerified

Project API keys in Langfuse can now be created with a permission level: Full access, Viewer, Ingestion (create traces, scores and media only) or Scores ingestion. Existing keys keep their current access.

Bears on: Observability / auditability

View source
·MCP / tool calling / APIVerified

Langfuse introduced stable, ID-based public APIs for managing LLM-as-a-Judge and code evaluators. Users can now programmatically create, version, and manage both evaluators and their associated evaluation rules.

Bears on: Observability / auditability

View source
View all 9 Langfuse updates →Tracked since Jun 2026 · Verified from public vendor sources

Pricing

Free Hobby (50K units/mo) · Core $29/mo · open source self host

usage

Free tier

Included quota

A billable unit is one trace, observation, or score. Hobby: 50K units a month, 2 users. Core: 100K units, unlimited users, 90 day data access, 4K req/min ingestion. Pro: 3 year retention, 20K req/min, compliance certifications. All paid plans share graduated $8 per 100K unit overage.

What is public

Langfuse publishes complete self serve pricing: free Hobby at 50K units a month, Core $29/mo, Pro $199/mo, Enterprise $2,499/mo, uniform $8 per 100K overage, and a free MIT licensed self host path with all product features open sourced since June 2025.

Billing mechanics

Flat plan fee plus graduated usage overage on billable units (traces, observations, scores). Tokens are tracked for LLM cost analysis but do not count as billable units. Unlimited users on all paid tiers.

Cost watchouts

Each trace, observation and score draws on the unit allowance, so multi-step agents use several units per request; Enterprise SSO, SSO enforcement and fine-grained RBAC sit behind the $300 a month Teams add-on on top of Pro.

Variable cost rationale

Unit metering scales with trace volume, but graduated overage at $8 per 100K units keeps cost growth predictable and there are no per seat fees.

Additional watchouts

Buyers should watch post acquisition roadmap and licensing signals under ClickHouse ownership, though all public commitments to open source have held since the January 2026 close.

Sales call required

No, self serve available

Free / trial

Free Hobby plan, no card (50K units/mo, 2 users)

Lowest paid plan

Core $29/mo (100K units, unlimited users, 90 day data access)

Commercial notes

Langfuse has joined ClickHouse and continues as its own product, with an MIT-licensed core for self-hosting and paid Enterprise additions. Its homepage cites use by 21 of the Fortune 50 and about 35,000 GitHub stars.

Key ambiguities

Enterprise custom volume pricing requires annual commitment. Self hosting is free on the MIT core but ClickHouse operations are real work.

Missing data

Enterprise custom volume rates unpublished.

Agentic Index verified 2026-09-21

Alternatives to Langfuse

The closest documented capability profiles to Langfuse among agent infrastructure platforms tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.

Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded

Contact us

Found a vendor we missed? Have feedback on the index? We'd love to hear from you.