Langfuse

Langfuse
Langfuse
  • Headquarters

    Berlin, Germany
  • Founded

    2022
  • CEO / Founder

    Clemens Rawert
  • Employees

    11-50
  • Funding

    Acquired — $4M
  • Products

    1 product listed
  • Avg. AI Score

    8.0/10

Profile updated

Products
1
Avg AI Score
8.0/10
Founded
2022
Funding
$4M
About Langfuse

Langfuse is an open-source LLM engineering platform founded in 2022 and headquartered in Berlin, Germany. The company was co-founded by Clemens Rawert and Marc Klingen to provide observability and development tooling for teams building applications on top of large language models.

The platform covers the full LLM application development lifecycle, including distributed tracing to debug and monitor model calls, prompt management with versioning, evaluation pipelines to assess output quality, and analytics dashboards for cost and latency tracking. Langfuse is available as a self-hosted open-source deployment or as a managed cloud service, and integrates with frameworks such as LangChain, LlamaIndex, and OpenAI's SDK.

Langfuse raised a Seed round of approximately $4 million, with backing from investors including Y Combinator, which the company participated in during the 2023 batch. The project's GitHub repository has accumulated substantial community adoption, reflecting demand for open-source observability tooling in the LLM application space. The company operates with a small team of under 50 employees.

Specialties
LLM ObservabilityPrompt ManagementAI Application TracingEvaluation PipelinesOpen-Source Developer ToolsLLM Analytics
Financials

Dated, source-cited figures. Estimates are marked.

Total Raised
$4M
langfuse.com
Products by Langfuse

All products by Langfuse reviewed by our AI panel of experts.

How Our Review Panel Scores Langfuse

Six independent reviewer personalities assessed Langfuse's product on their own terms. They landed within 0.6 points of each other, an unusually strong consensus.

The Decision Maker
8.3
The Domain Strategist
8.2
The Finance Lead
8.0
The Domain Practitioner
8.0
The Power User
7.9
The Skeptic
7.7

Averaged across 1 reviewed product in our catalog. Scores are out of 10.

Competitive Landscape

Who Langfuse competes with, and how they differ.

LangSmith logo
LangSmith8.1/10

LangChain's tracing, prompt-management, and eval platform overlaps almost feature-for-feature with Langfuse, but it is tightly coupled to the LangChain/LangGraph ecosystem while Langfuse is framework-agnostic and OSS-first.

2 products reviewed on TopReviewed
Helicone logo
Helicone7.9/10

Closest OSS-first peer: a proxy-based (one-line integration) LLM observability tool versus Langfuse's SDK/OpenTelemetry-based tracing, with a smaller evaluation and prompt-management feature set.

1 product reviewed on TopReviewed
Arize AI (Phoenix) logo
Arize AI (Phoenix)8.1/10

Open-source LLM tracing and evaluation rooted in Arize's traditional ML-monitoring product; its paid Arize AX tier targets larger enterprise observability budgets more aggressively than Langfuse's self-serve model.

1 product reviewed on TopReviewed
Braintrust logo
Braintrust8.0/10

Enterprise-focused LLM evaluation and observability platform with no self-hosted open-source option, relying on a higher-touch, sales-led go-to-market versus Langfuse's open-source, self-serve motion.

1 product reviewed on TopReviewed
Weights & Biases (Weave) logo
Weights & Biases (Weave)8.0/10

Extends an established ML experiment-tracking platform into LLM tracing/eval via Weave, leaning on incumbent ML-team relationships rather than Langfuse's LLM-native, OSS-first design.

1 product reviewed on TopReviewed
Datadog LLM Observability logo
Datadog LLM Observability7.0/10

Bundles LLM tracing and evaluation into its broader enterprise APM/observability suite, selling to existing Datadog customers on usage-based pricing rather than serving LLM-native or self-hosted developers directly.

1 product reviewed on TopReviewed
Company Timeline

Key milestones in Langfuse's history, with sources.

  1. Launch

    Langfuse v4 — observations-centric ClickHouse data model

    Langfuse shifted to a single wide, mostly immutable ClickHouse table that eliminates joins and read-time deduplication, cutting dashboard load times by at least 10x after 19x growth in data processed over the prior year.

    langfuse.com
  2. Acquisition

    Acquired by ClickHouse

    ClickHouse acquired Langfuse (terms undisclosed) alongside the close of its $400M Series D; the Langfuse team joined ClickHouse while the product continues operating as open-source (MIT license) and self-hostable, with Langfuse Cloud unchanged.

    clickhouse.com
  3. Milestone

    Open-sourced all product features under MIT license

    Langfuse moved LLM-as-a-Judge evaluations, annotation queues, prompt experiments, and the playground from its commercial tier to the open-source MIT license, leaving only enterprise security/platform features (SCIM, audit logs, data retention) commercial.

    langfuse.com
  4. Milestone

    Shipped Langfuse v3 — migrated core infrastructure from Postgres to ClickHouse

    The largest infrastructure upgrade in company history at the time, moving Langfuse from a Postgres-backed prototype to a ClickHouse-backed platform able to ingest hundreds of millions of records.

    langfuse.com
  5. Launch

    Launch Week 1 — rebranded as "The AI Engineering Platform"

    Langfuse introduced the LLM Playground, Datasets, and LLM-as-Judge evaluations, expanding beyond observability and prompt management; the launch was featured a second time as Product Hunt Product of the Day.

    langfuse.com
  6. Funding

    Seed round — $4M led by Lightspeed Venture Partners and La Famiglia

    Langfuse raised a $4M seed round led by Lightspeed Venture Partners and La Famiglia, with participation from Y Combinator and angel investors.

    langfuse.com
  7. Launch

    Public launch as "Product Analytics for LLM Apps"

    Langfuse launched publicly via Show HN, Y Combinator's Launch YC, and Product Hunt, where it won Product of the Day and a Golden Kitty Award.

    langfuse.com
  8. Founded

    Langfuse founded in Berlin

    Marc Klingen, Max Deichmann, and Clemens Rawert began working together in Berlin in 2022, initially building a usage-based billing product (Finto) before pivoting to LLM application tooling.

    langfuse.com

Explore AI Software Reviews

Browse multi-perspective AI panel reviews across hundreds of AI tools, agents, and platforms. Find the right software with insights from CTO, Developer, Marketer, Finance, and User perspectives.

Langfuse — Products & AI Reviews | TopReviewed.ai