2025

LangfuseWrapped

A year in review

2025 was a big year at 🪢 Langfuse.

We 🚢 more than ever.

The community is growing faster than ever... 📈

...and so are your applications 🚀

Let's take a l👀k back.

You all have been busy

Key metrics from our platform in 2025.

To the builders

Big thanks to all contributors of 2025.

⭐⭐⭐⭐

⭐⭐⭐⭐

Contributors

⭐⭐⭐⭐

Merged PRs

⭐⭐⭐⭐

GH Discussions

⭐⭐⭐⭐

GitHub Stars

A year of launches

We shipped a ton of new features this year.

Hold hover to ⭐ your favorite launches of the year

January

11 launches

⭐-Audit Logs ⭐-Track changes between prompt versions ⭐-Configure additional HTTP headers for LLM API keys ⭐-Dataset CSV upload ⭐-Commit Messages on Prompt Versions ⭐-Cmd+K Menu ⭐-Project-level Data Retention Policies ⭐-New API Reference ⭐-New Prompt Editor ⭐-JS/TS SDK supports trace sampling ⭐-OpenAI o3-mini support for playground and cost tracking

February

14 launches

⭐-Typed public API client in SDKs ⭐-Gemini 2.0 support for playground and cost tracking ⭐-Use JsonPath to select from Input, Output, or Metadata values in LLM as a Judge ⭐-Trace UI filtered view by log-level ⭐-Select all items in tables with a single click ⭐-OpenTelemetry Tracing Support ⭐-Open Source LLMOps Stack ⭐-Graph view for LangGraph traces ⭐-Model Context Protocol Prompt Server ⭐-Execute LLM-as-a-judge evaluations on existing data ⭐-Claude 3.7 Sonnet support ⭐-Google AI Studio support ⭐-Improved In-Product Onboarding ⭐-OpenAI GPT-4.5 support

March

14 launches

⭐-Langfuse Java Client ⭐-Delete dataset items and runs ⭐-Dataset items: Duplicate and add to many ⭐-Add support for WorkOS Identity Provider ⭐-Add support for Native Environments ⭐-Prompt Composability ⭐-Batch-export Scores via UI ⭐-Public API for Annotation Queues ⭐-OpenAI Response API support in SDKs ⭐-New Prompt View ⭐-New Trace View ⭐-Traces Table Peek View ⭐-Tool Calling and Structured Output in Playground ⭐-Playground support for Gemini 2.5 Pro Experimental

April

6 launches

⭐-Protected prompt labels ⭐-OpenAI 4.1 support for playground, evaluations and cost tracking ⭐-Admin APIs for Organization and Project Management ⭐-OpenAI o3 and o4-mini integration for playground, evaluations and cost tracking ⭐-Copy Docs as Markdown ⭐-Session Level Scores

May

11 launchesLaunch Week

⭐-Dataset Run Level Scores ⭐-Metrics API for custom analytics ⭐-Customize Product Module Visibility ⭐-Full Text Search ⭐-Save and share table views ⭐-Custom Dashboards ⭐-Claude Sonnet and Opus 4 support ⭐-Terraform Modules for AWS, Azure and GCP ⭐-OTEL-based Python SDK ⭐-Langfuse Evaluator Library ⭐-Pipecat Tracing Integration

June

9 launches

⭐-Open Sourcing all Product Features ⭐-Python SDK v3 is now Generally Available ⭐-OpenAI o3-pro and updated prices for o3 reflected in Langfuse cost tracking ⭐-Organize Your Prompts in Folders ⭐-CSV Download for Dashboard Charts ⭐-Exports of Datasets and Audit Logs ⭐-Improved JSON Handling in LangChain Prompts ⭐-Agentic Onboarding & Docs MCP Server ⭐-Histogram Charts in Custom Dashboards

July

13 launches

⭐-Pivot Tables in Custom Dashboards ⭐-Search to Navigate Large Prompt Collections ⭐-Message Placeholders in Chat Prompts ⭐-Gemini 2.5 Pro, Flash, and Flash Lite support for playground, evaluations, and cost tracking ⭐-n8n Node for Langfuse Prompt Management ⭐-Webhooks for Prompt Changes ⭐-AWS SDK Default Credential Provider Chain Support ⭐-Full-Text Search Across Prompt Content ⭐-Trigger Remote Custom Evaluations from the Langfuse UI ⭐-LiveKit Agents Tracing Integration ⭐-LLM Playground with Side-by-Side Comparison ⭐-Sessions in Annotation Queues ⭐-Slack Integration for Prompt Webhooks

August

8 launches

⭐-Annotation Queue Assignments ⭐-Docs now available as Markdown (.md) endpoints ⭐-OpenAI GPT-5 pricing now available in Langfuse ⭐-Additional provider options in LLM calls in playground and evals ⭐-Full-Text Search Across Dataset Items ⭐-New End-to-End Walkthrough Videos ⭐-Additional Observation Types for More Meaningful Span Context ⭐-TypeScript SDK v4 (GA)

September

4 launches

⭐-Experiment Runner SDK ⭐-Mutable Score Configs ⭐-Natural Language Filtering for Traces ⭐-Structured Output Support for Prompt Experiments

October

3 launches

⭐-Spend Alerts ⭐-LLM-as-a-Judge Execution Tracing & Enhanced Observability ⭐-Langchain v1 Support

November

15 launchesLaunch Week

⭐-Advanced Filtering for Public Traces and Observations API ⭐-Filter Sidebar for Tables ⭐-Amazon Bedrock AgentCore Integration ⭐-@Mentions and Reactions in Comments ⭐-IdP-Initiated SSO Support ⭐-Mixpanel integration ⭐-Langfuse for Agents ⭐-Annotation Support in Experiment Compare View ⭐-Baseline Support in Experiment Compare View ⭐-Filters in Compare View ⭐-Score Analytics with Multi-Score Comparison ⭐-Organize Your Datasets in Folders ⭐-JSON Schema Enforcement for Dataset Items ⭐-OpenAI GPT-5.1 support ⭐-Hosted MCP Server for Langfuse Prompt Management

December

6 launches

⭐-Pricing Tiers for Accurate Model Cost Tracking ⭐-Batch Add Observations to Datasets ⭐-OpenAI GPT-5.2 support ⭐-Dataset Item Versioning ⭐-v2 Metrics and Observations API (Beta)⭐-Filter Observations by Tool Calls and add Tool Calls to Dashboard Widgets

Powering the greatest

We couldn't be prouder to work with these companies.

⭐⭐⭐⭐

“We wanted something built for agents as users first. And that means API first. An agent should never get stuck waiting for a human because of a deficiency in the API. This is what Langfuse is.”

David Traina

Data Platform at Ramp

⭐⭐⭐⭐

“You can connect Claude Code to your Langfuse MCP - or the CLI directly - and it starts fixing and trying until it is fixed.”

Paolo Tamagnini

Senior Data Scientist at Trade Republic

⭐⭐⭐⭐

“Langfuse hits the sweet spot between engineering requirements and empowerment of non-technical users to contribute their domain expertise.”

Andreas Schuster

Head of Product, AI Help Experience at Canva

⭐⭐⭐⭐

“Building on Langfuse we saved 30% of external BPO cost by deflecting 50% of support conversations to AI.”

Ana Casado

Head of Operations Data and AI at SumUp

⭐⭐⭐⭐

“In just a few minutes you're up and running, and you're already able to see what's going on.”

Niels Rogge

Machine Learning Engineer at Hugging Face

⭐⭐⭐⭐

“Langfuse has enabled our developers to get extremely fast feedback. It's fundamental to how our developers understand their AI implementations.”

Walt Wells

Staff Software Engineer at Khan Academy

⭐⭐⭐⭐

And thousands more...

⭐⭐⭐⭐

“Langfuse enables us to track every prompt, response, cost, and latency in real time, turning black-box models into auditable, optimizable assets.”

Walid Mehanna

Chief Data & AI Officer at Merck

⭐⭐⭐⭐

“When a customer pings us, we instantly open Langfuse. This allows keeping support resolution under 8 minutes.”

Alexander Danilowicz

Co-founder at Magic Patterns

Thank you for being part of this journey

If there was one thing we could wish for...