MLflow
DatabricksActive 2026; Apache 2.0; observability/tracing for agents+LLMs+ML
Active 2026; Apache 2.0; observability/tracing for agents+LLMs+ML
Openness
5 high confidence- license
- Apache-2.0
- source
- public
- core-gated
- ungated
license:Apache-2.0;source:public;core-fully-OSS(Databricks-Managed-MLflow is third-party hosting, not gated core)
- https://github.com/mlflow/mlflow/blob/master/LICENSE.txt recorded 2026-06-04
flagship phase-C verification source
- https://github.com/mlflow/mlflow recorded 2026-06-04
flagship phase-C verification source
- https://mlflow.org/ recorded 2026-06-04
flagship phase-C verification source
Adoption
5 high confidenceRe-scored 4->5. 37.10M PyPI downloads in the last 30 days (verified on pypistats; the prior note already reported 25-30M+ but mislabeled the reach as 1M-10M). The map's prevailing usage_volume level-5 floor is >10M/mo (cf. pydantic-ai ~31M, langgraph ~58M at level 5). MLflow is the de facto open experiment-tracking / ML-lifecycle platform and clears the floor comfortably.
- https://pypistats.org/api/packages/mlflow/recent recorded 2026-06-25
last_month downloads = 37,102,636
- https://github.com/mlflow/mlflow recorded 2026-06-25
mlflow/mlflow repo (experiment-tracking platform) confirmed live
Capability
4 high confidenceSingle open platform combining distributed tracing, prompt versioning, automated evaluation, and trace replay for agents/LLMs plus classic ML tracking. No standard benchmark for observability; rated on feature matrix.
- https://mlflow.org/genai/ recorded 2026-06-04
flagship phase-C verification source
- https://mlflow.org/top-5-agent-observability-tools/ recorded 2026-06-04
flagship phase-C verification source
Unchanged since 2026-07-30 (last edited, not re-checked)