Registry › Compare › Langfuse vs MLflow

Langfuse vs MLflow

MLflow leads on trust: 89.7/100 (Grade A) against 79.9/100 (Grade B), a 9.8-point gap. Langfuse leads on transparency; MLflow leads on supply-chain integrity and provenance.

B Langfuse 79.9

🪢 Open source AI engineering platform: AI Gateway, agent observability, evals, and prompt/skill management, built on OpenTelemetry. One open platform for every AI call and every agent. Self-host or use Langfuse Cloud.

langfuse/langfuse · #196 overall · #6 Observability & Evaluation · coverage A (4/5)

Choose Langfuse if transparency matters most.

+1.6Transparency: OSSF Scorecard 7.4 against 5.5

A MLflow 89.7

The open source AI engineering platform for agents, LLMs, and ML models. MLflow enables teams of all sizes to debug, evaluate, monitor, and optimize production-quality AI applications while controlling costs and managing access to models and data.

mlflow/mlflow · #36 overall · #2 Observability & Evaluation · coverage A (4/5)

Choose MLflow if supply-chain integrity and provenance matter most.

+7.2Identity / Provenance: package provenance attested, against none
+5.2Safety / Integrity: 100% of recent commits signed, against 98%

Where they differ

14.2

Safety / IntegrityMLflow +5.2

19.4

10.8

Identity / ProvenanceMLflow +7.2

18.0

14.8

TransparencyLangfuse +1.6

13.2

19.9

MaintenanceMLflow +0.1

20.0

AdoptionLangfuse +0.3

19.7

+0.2

Runtime calibrationLangfuse +0.8

-0.6

Full evidence table

An independent, evidence-based trust comparison of Langfuse and MLflow, two Observability & Evaluation projects in the HVTracker registry. Scores come from public, checkable signals — supply-chain provenance, OSSF Scorecard, maintenance, and adoption — not popularity.

Full evidence

Signal	Langfuselangfuse/langfuse	MLflowmlflow/mlflow
HVTrust score	79.9	89.7
Evidence grade	B	A
Coverage grade	A	A
Overall rank	#196	#36
Rank in Observability & Evaluation	#6	#2
GitHub stars	37.4k	28.8k
Last updated	1d ago	today
Build provenance	No	Yes
OSSF Scorecard	7.4 / 10	5.5 / 10
License	NOASSERTION	Apache-2.0
Downloads	10.0M/wk	5.2M/wk
Trust dimensions (points earned)
Safety / integrity / 25	14.2	19.4
Identity & provenance / 18	10.8	18.0
Transparency / 17	14.8	13.2
Maintenance / 20	19.9	20.0
Adoption / 20	20.0	19.7
Runtime capability surface (full matrix)
MCP server	Implemented	Implemented
External providers	2 — Anthropic, OpenAI	3 — Amazon Bedrock, Anthropic, Postgres
Requires API keys	Yes	No
Plugin surface	plugins	plugins
Provenance drift	Unknown	Unknown

Open in the live compare tool → Langfuse profile MLflow profile More Observability & Evaluation →

How to read this: HVTrust (0–100) weighs supply-chain signals (provenance, OSSF Scorecard, signed commits, open license) alongside real-world adoption, scaled by an evidence-confidence factor. Grade bands: A ≥ 80, B ≥ 65, C ≥ 50, D < 50. Signals refresh daily. Full methodology v4.4 →