HTTP 200 verified twice
AI infrastructure · Tool
Langfuse
ResearchedLangfuse is an open-source AI engineering platform (by ClickHouse) for tracing, evaluating, and improving LLM applications. It offers LLM observability, prompt management, evaluation, and metrics, processing 10+ billion observations monthly.
Official site snapshots
Visit live site See the official site at a glance
Read-only public captures of Langfuse’s homepage. Screenshots are dated, never live embeds, and open full-screen.
In one minute
Start here for the decision-making essentials: what Langfuse does, who it is for, how it is accessed, and the first-party sources behind this profile.
Pricing5 options
Free tier available ('Start free')
managed cloud launched via 'Launch App'.
Hobby plan is Free with no credit card required, 50k units/month, 30 days data access, 2 u
Core plan is $29/month with 100k units/month included (additional $8/100k units, lower wit
Pro plan is $199/month with 3 years data access, data retention management, unlimited anno
Platforms
AI engineering platformLangfuse Cloud Service
API accessNot public
Founded2023
AvailabilityWeb / remote
LicenseMIT
Best suited to
Source-backed fit Teams collaboratively building LLM applications Engineers needing production LLM observability and tracing Organizations requiring self-hostable AI infrastructure Enterprises needing SOC2/ISO27001/HIPAA-ready LLM tooling Builders integrating OpenAI, LangChain, or other LLM frameworks Used by 19 of Fortune 50 companies and 100,000+ engineers.
Decision support
6 sourced answersCommon questions and adoption checks
Short answers to the questions buyers and builders commonly ask about Langfuse. Each answer cites the shared ledger below, where every source is listed once.
01What does Langfuse say it can do?
Trace, evaluate, and improve AI agents with one open platform, connecting tracing, monitor · Observability with hierarchical traces capturing every LLM call, tool invocation, and retr · Product areas include LLM Observability, Prompt Management, Evaluation, and Metrics. · Observability and tracing for LLM applications
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.
02Who is Langfuse intended for?
Used by 19 of Fortune 50 companies and 100,000+ engineers. · Trusted by 40,000+ builders. · Teams collaboratively developing LLM applications
Used by 19 of Fortune 50 10+ billion observations/month 100,000+ engineers building on Langfuse
03What use cases does Langfuse describe?
LLM Observability, Prompt Management, and Evaluation. · Production-scale LLM engineering platform processing 10+ billion observations/month. · Debugging, analyzing, and iterating on LLM applications · RAG observability and evaluations, including retrieval relevance, answer faithfulness, and
Product Overview LLM Observability Prompt Management Evaluation Metrics
04What should teams verify before adopting Langfuse?
Example project provides view-only access · Image generation limited to 3 generations per minute
The example project provides view-only access .
Ledger citation[11] langfuse.com/docs/demo
05What pricing information is available for Langfuse?
Free tier available ('Start free'); managed cloud launched via 'Launch App'. · Hobby plan is Free with no credit card required, 50k units/month, 30 days data access, 2 u · Core plan is $29/month with 100k units/month included (additional $8/100k units, lower wit · Pro plan is $199/month with 3 years data access, data retention management, unlimited anno
Start free S Documentation D Read story Read story Read story Read story
06Does Langfuse document API access?
API-first architecture · Exposes custom evaluation workflows via API/SDKs.
Platform API-first architecture
Decision guide
AI infrastructureCapabilities and operating fit
This profile connects the jobs Langfuse is described as handling with its delivery model, access options and the subjects used to match it to related products in this directory.
Common use cases
- LLM observability and tracing for AI agents
- Debugging and analyzing LLM application behavior
- Prompt management with versioning and deployment
- Evaluation of output quality and production health
- Monitoring cost, latency, and session/user metadata
- Managing datasets and experiments for LLM apps
Topics mapped
Verified capabilities
- Trace, evaluate, and improve AI agents with one open platform, connecting tracing, monitor
- Observability with hierarchical traces capturing every LLM call, tool invocation, and retr
- Product areas include LLM Observability, Prompt Management, Evaluation, and Metrics.
- Observability and tracing for LLM applications
- Prompt Management with version control and deployment
- Evaluation to measure output quality and monitor production health
- Trace capture including LLM, retrieval, embedding, and API calls; multi-turn session track
- LLM Tracing
Recorded integrations
Supports ingesting traces via OpenAI, LangChain, or the SDKs.Native SDKs for Python and JS plus 100+ library/framework integrationsOpenAI SDKLangChain SDKBuilt on OpenTelemetry, with Python SDK, JS/TS SDK, CLI, MCP Server, and an OpenTelemetry Direct integrations with AI model providers including OpenAI, Anthropic, Amazon Bedrock, GConnects through API gateways and proxies including OpenRouter, Portkey, LiteLLM Proxy, HeIntegrates with no-code agent builders and tools such as Dify, Flowise, Langflow, n8n, OpeIntegrates with analytics tools that can visualize Langfuse traces and metrics, including Integrates with developer tools including GitHub Copilot, Cursor, Claude Code, Codex, VS C
Intended audiences
Access signals
- Pricing model
- Free tier available ('Start free'); managed cloud launched via 'Launch App'. · Hobby plan is Free with no credit card required, 50k units/month, 30 days data access, 2 u · Core plan is $29/month with 100k units/month included (additional $8/100k units, lower wit · Pro plan is $199/month with 3 years data access, data retention management, unlimited anno
- API
- Not publicly listed
- Source links
- 17 recorded
Source-backed
Updated July 28, 2026Verified facts
Each fact points to a recorded source, making it easy to distinguish verified product information from claims that need checking.
Langfuse
Open-source AI engineering platform · Operated by/part of ClickHouse · Trace, evaluate, and improve AI agents
Yes
Trace, evaluate, and improve AI agents with one open platform, connecting tracing, monitoring, datasets, experiments, and evaluation.
Langfuse is operated by ClickHouse ('Langfuse joins ClickHouse').
LLM Observability, Prompt Management, and Evaluation.
Used by 19 of Fortune 50 companies and 100,000+ engineers.
View 32 more verified facts
Self-hosting supported via Docker Compose, Kubernetes (Helm), AWS (Terraform), GCP (Terraform), and Azure (Terraform); or use the managed 'Launch App' cloud offering.
Supports ingesting traces via OpenAI, LangChain, or the SDKs.
Observability with hierarchical traces capturing every LLM call, tool invocation, and retrieval step; filter by user, session, cost, latency, or custom metadata.
Free tier available ('Start free'); managed cloud launched via 'Launch App'.
Community support via GitHub (31.6k Stars, 300+ Contributors), Q&A threads, and roadmap threads.
Production-scale LLM engineering platform processing 10+ billion observations/month.
Langfuse is part of ClickHouse ("Langfuse by ClickHouse", "Langfuse joins ClickHouse").
Product areas include LLM Observability, Prompt Management, Evaluation, and Metrics.
Hobby plan is Free with no credit card required, 50k units/month, 30 days data access, 2 users, community support via GitHub.
Core plan is $29/month with 100k units/month included (additional $8/100k units, lower with volume), 90 days data access, unlimited users, in-app support.
Pro plan is $199/month with 3 years data access, data retention management, unlimited annotation queues, high rate limits, SOC2 & ISO27001 reports, HIPAA-ready region, prioritized in-app support plus optional Teams Add-on $300/mo.
Enterprise plan is $2499/month with SSO (e.g. Okta), SSO enforcement, fine-grained RBAC, Dedicated Slack / MS Teams Channel, Audit Logs, SCIM API, custom rate limits, Uptime SLA, Support SLA, dedicated support engineer, yearly commitment op
Langfuse has an open-source codebase with 31.6k GitHub stars and 300+ contributors.
Self-hosting guides are available for Docker Compose, Kubernetes (Helm), AWS (Terraform), GCP (Terraform), and Azure (Terraform).
Pro tier includes SOC2 & ISO27001 reports and a HIPAA-ready region.
Enterprise tier adds Enterprise SSO (e.g. Okta), SSO enforcement, fine-grained RBAC, Audit Logs, SCIM API, Uptime SLA, and Support SLA.
Trusted by 40,000+ builders.
Support options range from community Q&A (1.8k threads) and GitHub community on Hobby, in-app support on Core, prioritized in-app support plus optional Teams Add-on on Pro, to dedicated support engineer and SLA-backed channels on Enterprise
true
AI engineering platform
Debugging, analyzing, and iterating on LLM applications
Teams collaboratively developing LLM applications
Self-hostable
Observability and tracing for LLM applications
Prompt Management with version control and deployment
Evaluation to measure output quality and monitor production health
Native SDKs for Python and JS plus 100+ library/framework integrations
OpenTelemetry-based for compatibility
API-first architecture
Data exports to blob storage
Enterprise security and administration features
Trace capture including LLM, retrieval, embedding, and API calls; multi-turn session tracking; user tracking; agent graph representation
Practical capabilities
8 documented areasWhat it helps with
A concise view of the jobs, capabilities and integrations described in the recorded product sources.
LLM observability and tracing for AI agents
Debugging and analyzing LLM application behavior
Prompt management with versioning and deployment
Evaluation of output quality and production health
Monitoring cost, latency, and session/user metadata
Managing datasets and experiments for LLM apps
LLM Observability, Prompt Management, and Evaluation.
Production-scale LLM engineering platform processing 10+ billion observations/month.
Availability
Source checkedWhere it runs and where to get it
Documented product formats, platforms and official distribution destinations. Availability can vary by region and plan.
Cost / license
Free tier available ('Start free'); managed cloud launched via 'Launch App'. · Hobby plan is Free with no credit card required, 50k units/month, 30 days data access, 2 u · Core plan is $29/month with 100k units/month included (additional $8/100k units, lower wit · Pro plan is $199/month with 3 years data access, data retention management, unlimited annoMITYesLangfuse has an open-source codebase with 31.6k GitHub stars and 300+ contributors.trueMIT · Yes · Langfuse has an open-source codebase with 31.6k GitHub stars and 300+ contributors. · true
Origin
Langfuse joined ClickHouse
Platforms
AI engineering platformLangfuse Cloud ServiceLive, shared web example project viewable before signup
Adoption notes
DeploymentSelf-hosting supported via Docker Compose, Kubernetes (Helm), AWS (Terraform), GCP (Terraf · Self-hosting guides are available for Docker Compose, Kubernetes (Helm), AWS (Terraform), · Self-hostable · Self-hosting via Docker Compose, Kubernetes (Helm), AWS (Terraform), GCP (Terraform), and
LicenseMIT · Yes · Langfuse has an open-source codebase with 31.6k GitHub stars and 300+ contributors. · true
Model supportNot disclosed by source
Data controlPro tier includes SOC2 & ISO27001 reports and a HIPAA-ready region. · Enterprise tier adds Enterprise SSO (e.g. Okta), SSO enforcement, fine-grained RBAC, Audit · Enterprise security and administration features
Learning curveIntermediate
Primary use casesLLM observability and tracing for AI agents, Debugging and analyzing LLM application behavior, Prompt management with versioning and deployment, Evaluation of output quality and production health, Monitoring cost, latency, and session/user metadata, Managing datasets and experiments for LLM apps, LLM Observability, Prompt Management, and Evaluation., Production-scale LLM engineering platform processing 10+ billion observations/month., Debugging, analyzing, and iterating on LLM applications, RAG observability and evaluations, including retrieval relevance, answer faithfulness, and, Agent evaluation, assessing end-to-end trajectories for agents with tool use and planning, trace, evaluate, and improve LLM applications and AI agents in production, Inspecting traces that show timing, costs, input/output, and evaluation scores, Scoring live traces in production to turn interesting examples into datasets and compare c, Blocking deploys on regressions through CI/CD experiments., Running deterministic checks via Code Evaluators.
What to verify before adopting
- Example project provides view-only access
- Image generation limited to 3 generations per minute
Evolution and major updates
12 dated updatesLangfuse timeline
A concise history of software releases and material product changes. Events appear only when a dated source supports what changed.
Latest first · exact dates
Showing the newest updates and meaningful milestones. Open an entry for its summary and source.
Any table is a chart
Open detailsToggle the Observations table into a chart using the same query and filters, with chart type, metric, aggregation, and breakdown options, then add it to a dashboard.
View source [13]Run Experiments on Versioned Datasets
Open detailsFetch datasets at specific version timestamps and run experiments on historical dataset versions via UI, API, and SDKs for full reproducibility.
View source [7]Corrected Outputs for Traces and Observations
Open detailsCapture improved versions of LLM outputs directly in trace views to build fine-tuning datasets and drive continuous improvement with domain expert feedback.
View source [7]Inline Comments on Observation I/O
Open detailsAnchor comments to specific text selections within trace and observation input, output, and metadata fields.
View source [7]Filter Observations by Tool Calls and add Tool Calls to Dashboard Widgets
Open detailsAdd filtering, table columns, and dashboard widgets for analyzing tool usage in LLM applications.
View source [7]OpenAI GPT-5.2 support
Open detailsAdds day 1 support for OpenAI GPT-5.2 across all major Langfuse features.
View source [7]OpenAI GPT-5.2 support
Open detailsLangfuse now supports OpenAI GPT-5.2 with day 1 support across all major features.
View source [7]
View 5 more dated releasesFull source-backed history
Citation ledger
19 unique pagesRecorded sources
Facts, answers, structured details, milestones and primary resource links cite this shared ledger. Each external page appears once; release tags from the same GitHub project are grouped under one release history.
- 1langfuse.com 15 facts · 4 answers · Official site
- 2langfuse.com/docs 14 facts · 4 answers · Documentation
- 3langfuse.com/pricing 12 facts · 3 answers · Pricing
- 4langfuse.com/guides 12 facts · 2 answers · Official site
- 5langfuse.com/press 12 facts · 1 answer · Official site
- 6langfuse.com/integrations 11 facts · Official site
- 7langfuse.com/changelog 11 milestones
- 8langfuse.com/privacy 10 facts · Security
- 9langfuse.com/about 9 facts · Official site
- 10langfuse.com/blog 8 facts · Official site
- 11langfuse.com/docs/demo 1 fact · 2 answers · Documentation
- 12langfuse.com/docs/evaluation/overview 1 answer · Documentation
- 13langfuse.com/changelog/2026-07-24-any-table-is-a-chart 1 milestone
- 14langfuse.com/docs/metrics/overview Documentation
- 15langfuse.com/docs/observability/get-started Documentation
- 16langfuse.com/docs/observability/overview Documentation
- 17langfuse.com/docs/prompt-management/features/playground Documentation
- 18langfuse.com/docs/prompt-management/get-started Documentation
- 19langfuse.com/docs/roadmap Documentation
Research status120 substantive facts · 17 source pages · quality score 100/100
Connected directory
Semantic matchesExplore related solutions
These tools share a workflow, capability or audience with Langfuse. They may complement it rather than replace it, and are not presented as integrations or endorsements.
