mlflow
28,924Open-source platform for the machine learning lifecycle — experiments, tracking, models, and LLM evaluation.
Activity band details
Activity band A to F: a bin of a score that adds GitHub stars (log scale), a push to the upstream repository in the 12 months before `as_of`, the number of declared integration surfaces and, where the catalogue records them, citing papers and benchmark evaluations. It measures popularity and activity. It is not a level of evidence: no band means a paper cites the tool or that it was measured. Recency is judged as of 2026-11-22.
Computed band: C
(raw score 5.28).
Shown band: C
.
Facets:
documented_in=0evaluated_on=0integration_surfaces=1.5paper_citations=0recent_push=2stars=1.78
Source of truth: pipeline/maturity/score.py
Description
MLflow is the open-source ML lifecycle platform stewarded by Databricks. Recent LLM-focused additions (MLflow 2.x) bring LLM tracing (OpenTelemetry), LLM evaluation, and a deployments registry — making it usable as an observability + eval backbone for LLM applications alongside classic ML.
Why it's included
Represents the 'classical-MLOps-extended-to-LLMs' point: teams with MLflow infrastructure can adopt LLM evaluation without a greenfield tool. Differentiated from arize-phoenix by its ML lifecycle lineage and Databricks ecosystem integration.
Addresses failure modes
These tags are the catalog's claim about what mlflow is built to catch or prevent. A tag is not a test result. SAICA-Bench has not measured mlflow, so none of these tags is tested here.
- Context pollution — Trajectory drift, hallucination spirals, overlong or off-task context.
Records evidence of
mlflow can record evidence of these failures (traces, or checks you configure yourself), but it does not detect or prevent them on its own. They are not counted as failure modes it addresses.