LangFuse
Open source agent evals and observability platform
Ranked #1 in Self-hosted LLM observability tools, and on 1 more ranking.
Langfuse is an open-source AI engineering platform for observability and evaluation. It enables teams to trace, evaluate, and improve AI agents with one integrated system.
At a glance
| Best for | AI engineering teams building production-grade LLM applications |
From LangFuse's own website.
Company facts
| Social profiles | LinkedIn, X, YouTube |
Facilities and features
Its website also mentions:
- Published pricing
- Free plan
- Self-serve sign-up
- API
- Integrations
- SOC 2
- GDPR
What stands out
The AI tool captures hierarchical traces of every LLM call and supports multiple evaluation methods.
“Hierarchical traces capture every LLM call, tool invocation, and retrieval step.”LangFuse, own website
The AI tool offers over 100 integrations with major frameworks and model providers without lock-in.
“Langfuse works with any language and framework supporting OTel instrumentation.”LangFuse, own website
How it ranks
LangFuse is ranked 1st of 10 in Self-hosted LLM observability tools (Lister Score 92.0) and 1st of 10 in Best LLM observability tools (Lister Score 92.0).
#1 in Self-hosted LLM observability tools · Lister Score 92.0
The AI tool captures hierarchical traces of every LLM call and supports multiple evaluation methods.
How the score is made up: fit to this ranking's criteria 40; evidence on its own site 20; how complete its published information is 16; buying information on its own site 16.
“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.”LangFuse, own website
“Self-host at scale Docker Compose Kubernetes (Helm) AWS (Terraform) GCP (Terraform) Azure (Terraform)”LangFuse, own website
| Rank | Self-hosted LLM observability tools | Score |
|---|---|---|
| 1 | LangFuse | 92.0 |
| 2 | Morph | 92.0 |
| 3 | Manifest | 90.0 |
| 4 | Shakudo | 90.0 |
| 5 | Entelligence AI | 88.0 |
| 6 | Fiddler AI | 88.0 |
| 7 | axem | 84.0 |
| 8 | Iterate.ai | 84.0 |
| 9 | MLflow AI Platform | 82.0 |
| 10 | PromptLayer | 82.0 |
#1 in Best LLM observability tools · Lister Score 92.0
Langfuse is an open-source platform for agent evaluations and observability. It enables teams to trace, evaluate, and improve AI agents efficiently.
The AI tool captures hierarchical traces of every LLM call and tool invocation.
How the score is made up: fit to this ranking's criteria 40; evidence on its own site 20; how complete its published information is 16; buying information on its own site 16.
“Open Source Agent Evals & Observability Trace, evaluate, and improve AI agents with one open platform.”LangFuse, own website
| Rank | Best LLM observability tools | Score |
|---|---|---|
| 1 | LangFuse | 92.0 |
| 2 | AgentOps | 88.0 |
| 3 | Arize AI | 88.0 |
| 4 | Fiddler AI | 88.0 |
| 5 | Weights and Biases | 88.0 |
| 6 | HoneyHive | 84.0 |
| 7 | MLflow AI Platform | 82.0 |
| 8 | PromptLayer | 82.0 |
| 9 | Dynamiq | 80.0 |
| 10 | FloTorch | 76.0 |
Frequently asked questions
How does LangFuse rank on Lister?
LangFuse is ranked 1st of 10 in Self-hosted LLM observability tools (Lister Score 92.0) and 1st of 10 in Best LLM observability tools (Lister Score 92.0).