langfuse
Langfuse Python SDK - LLM observability/tracing, datasets, experiments, LLM-as-a-judge evaluation, and prompt management
Decision gist · record as of 2026-08-14
Yes. Langfuse is actively maintained, has low install friction, carries a permissive MIT license, and ranks in the top 1000 PyPI packages. It addresses a real need for LLM observability and evaluation. The v4 rewrite is recent; ensure your use case aligns with the current API by consulting the migration guide if upgrading from v3.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.10 or later; environment variables LANGFUSE_PUBLIC_KEY, LANGFUSE_SECRET_KEY, and LANGFUSE_BASE_URL must be set to connect to a Langfuse instance.
- Low install friction with a pure-wheel distribution.
- Active maintenance with a recent release (3 days old).
License · maintenance · safety
MIT (permissive) — MIT license permits commercial and private use with minimal restrictions, making it suitable for most production deployments.
last release 2026-08-11 (3 days)
0 known vulnerabilities (OSV.dev, 2026-08-14) · 25,407,481 downloads/mo, #903 on PyPI
Alternatives
Verify before relying
pip install langfuse
from langfuse import get_client
langfuse = get_client()
with langfuse.start_as_current_observation(as_type="span", name="process") as span:
span.update(output="Done")
langfuse.flush()- Whether OpenAI and LangChain integrations are automatically available or require additional setup steps.
- Performance overhead of tracing in production workloads with high request volumes.
- Backward compatibility guarantees for v4 SDK beyond the migration guide.
What it is and what it does
Langfuse is a comprehensive observability and evaluation platform for LLM applications, delivered as a Python SDK. It provides OpenTelemetry-based tracing to capture spans and generations from your LLM workflows, with built-in integrations for OpenAI and LangChain. Beyond tracing, it supports offline evaluation through datasets and experiments, allowing you to test prompt and model changes with regression testing and CI integration via GitHub Actions. The SDK also includes LLM-as-a-judge evaluation, custom scoring, and prompt management capabilities.
The SDK depends on httpx for HTTP requests, pydantic for data validation, backoff for retry logic, wrapt for instrumentation, packaging for version handling, and OpenTelemetry libraries (opentelemetry-api, opentelemetry-sdk, opentelemetry-exporter-otlp-proto-http) for standards-based tracing. It requires Python 3.10 or later and is actively maintained, with a v4 rewrite released in March 2026 that introduced breaking changes documented in a migration guide.
Use it for
- Trace LLM calls and nested operations in production to debug latency, errors, and model behavior.
- Run offline experiments to evaluate prompt or model changes before deploying to production.
- Implement custom evaluation logic and LLM-as-a-judge scoring for automated quality assessment.
- Centralize prompt management and versioning across multiple applications.
- Set up CI/CD regression testing for LLM applications using datasets and GitHub Actions.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes.
Langfuse is actively maintained, has low install friction, carries a permissive MIT license, and ranks in the top 1000 PyPI packages. It addresses a real need for LLM observability and evaluation. The v4 rewrite is recent; ensure your use case aligns with the current API by consulting the migration guide if upgrading from v3.
Install
langfuse on PyPI
Before you install
Low install friction with a pure-wheel distribution. Active maintenance with a recent release (3 days old). Requires Python 3.10 or later.
Requires Python 3.10 or later; environment variables LANGFUSE_PUBLIC_KEY, LANGFUSE_SECRET_KEY, and LANGFUSE_BASE_URL must be set to connect to a Langfuse instance.
License in practice
MIT license permits commercial and private use with minimal restrictions, making it suitable for most production deployments.
Quickstart
pip install langfuse
from langfuse import get_client
langfuse = get_client()
with langfuse.start_as_current_observation(as_type="span", name="process") as span:
span.update(output="Done")
langfuse.flush()Verify before relying
- Whether OpenAI and LangChain integrations are automatically available or require additional setup steps.
- Performance overhead of tracing in production workloads with high request volumes.
- Backward compatibility guarantees for v4 SDK beyond the migration guide.
Package facts
| License | MIT permissive |
| Python support | Supports the current Python release <4.0,>=3.10 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 8 packageshttpxpydanticbackoffwraptpackagingopentelemetry-apiopentelemetry-sdkopentelemetry-exporter-otlp-proto-http |
| Maintenance | Actively maintained 3 days since the last release |
| First released | |
| Downloads | 25,407,481 / month, #903 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
Evidence: langfuse-4.14.4-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “prompt management sdk”
- langfuseLangfuse is a Python SDK for observability, tracing, and evaluation…
- vellum-aiVellum is an SDK for building and deploying production-grade AI…
- promptlayerPromptLayer is a Python SDK for managing, testing, and monitoring LLM…
Give your agent the search over MCP, or paste the wish link into any chat.
More Monitoring packages
Wraps any iterable to display a real-time progress bar in the terminal or Jupyter notebook, showing iteration count, elapsed time, and estimated time remaining.
Provides generated Python code for OpenTelemetry semantic conventions, enabling standardized attribute naming and constant definitions for instrumentation and telemetry collection.
Install it if you are using OpenTelemetry and want to follow semantic conventions correctly.
Provides the reference implementation of the OpenTelemetry API for collecting and exporting traces, metrics, and logs from Python applications.
Provides the abstract API and interfaces for OpenTelemetry instrumentation in Python, defining how to emit traces, metrics, and logs without tying code to a specific SDK implementation.
Exports OpenTelemetry observability data to an OpenTelemetry Collector using Protobuf-encoded messages over HTTP.
Install it if you are using OpenTelemetry in Python and need to send data to a Collector over HTTP.
Provides automatic instrumentation commands and programmatic APIs to inject distributed tracing into Python applications without code changes, detecting and instrumenting packages used by your program.
Install it if you need distributed tracing without code changes and have compatible instrumented packages in your environment.
See also opik · opentelemetry-instrumentation-langchain · deepeval · arize · promptflow-devkit · traceloop-sdk · promptflow-tracing · openlit · arize-phoenix · langwatch