Start free

Langfuse connector · LLM cost and trace reports

Read & reportRead-only source

See what your LLM app costs with Langfuse.

Bring Langfuse daily metrics, traces and evaluation scores into your reports and AI assistants, with trace counts, cost, latency and users.

  • Track daily traces and model cost
  • Find the slowest and most expensive traces
  • Ask ChatGPT or Claude how scores moved this month
LangfuseSheetsSheetsData StudioData StudioChatGPTChatGPTClaudeClaude

Start free Explore the data

Updated November 2026

Langfuse reportSample data
DateTracesObservations
2026-11-124,21018,900
2026-11-133,98017,400
2026-11-144,56020,100
Refreshes on your schedule in Google Sheets

Langfuse reporting for one project

One Langfuse project is the data source.

Daily metrics

Read traces, observations and total cost for each day in the report period.

Traces

List traces in the report period with name, user, latency, cost and date.

Scores

See evaluation scores with name, value, trace and date.

Cost reviews

Compare cost day by day and by user.

Report datasets and details

Reports you can build

4 reporting datasets, each with its own fields. Pick one in the app, then choose your columns.

  • Langfuse LLM traces & metrics
  • Daily LLM metrics
  • Traces
  • Evaluation scores
What each dataset reports
  • Langfuse LLM traces & metrics

    Daily trace counts, observation counts, scores, and model costs.

  • Daily LLM metrics

    Langfuse daily trace counts, observation counts, and total model cost.

  • Traces

    Langfuse LLM execution traces with name, user ID, latency, and cost.

  • Evaluation scores

    Langfuse evaluation scores linked to traces.

Metrics and dimensions

15 report fields across 3 datasets.

Search the available fields. Hover or tap a field to see what it means.

Send Langfuse data anywhere

Connect Langfuse to ChatGPT or Claude for LLM cost questions

Use Datapipe’s MCP access as your Langfuse AI integration. Ask about daily cost, slow traces and score trends.

Example

Which days cost the most this month, and which traces were slowest?

Sort this month’s daily metrics by total cost, then sort traces by latency.

Start with Langfuse

What you need

A Langfuse public and secret key from Project Settings, then API Keys. EU, US, Japan and HIPAA Cloud regions work.

What to know

Read-only. Traces, prompts and scores aren’t changed. Prompt and completion text isn’t included, and self-hosted Langfuse isn’t supported.

How it works

  1. Create your free account, then choose Langfuse in Data sources.

  2. Enter your public and secret keys, and your region’s address if it isn’t EU Cloud.

  3. Choose a report and destination: a sheet, Data Studio, ChatGPT or Claude.

Works well with Langfuse

Add both to the same project. Each keeps its own report, so you can check them side by side or ask ChatGPT or Claude a question that needs both.

Questions

Datapipe reads daily traces, observations and cost, traces with latency and cost, and evaluation scores.

No. Reports cover counts, cost, latency and scores, not the content of each call.

No. Langfuse is read-only in Datapipe.

EU, US, Japan and HIPAA Langfuse Cloud. Self-hosted Langfuse isn’t supported.

Free to start. Paid plans add more data sources and refresh faster.

Free and Starter refresh once a day. Pro refreshes every hour. Agency refreshes every 15 minutes on the first 50 data sources.

One for each Langfuse account you add to a project. Adding the same account to another project counts again. Reports and destinations do not use extra data sources.

Know what your LLM features cost.

Start free, add Langfuse and bring its metrics into the tools you already use.

Start free See setup requirements