MLflow: CreateModelVersion source validation does not check READ permission on referenced run_id
MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. Prior to 3.15.0, CreateModelVersion accepts a run_id or model_id after _validate_source_run() or _validate_source_model() in mlflow/server/handlers.py verifies only path containment, allowing authenticated users to create a model version that references another user's artifact directory and read files through GET /model-versions/get-artifact without the required READ permission. This issue is fixed in version 3.15.0.
describes (1)
Related CVEs.
| ID | severity | exploitable | ||
|---|---|---|---|---|
| CVE-2026-69148US | —OT elevated | — | ||
MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. Prior to 3.15.0, CreateModelVersion accepts a run_id or model_id af… | ||||
references
External citations — patches, advisories, analysis.
- https://github.com/mlflow/mlflow/commit/4bb7474771c3be808cd9e129defef9305f2869beWEB
- https://github.com/mlflow/mlflow/pull/24293WEB
- https://github.com/mlflow/mlflow/releases/tag/v3.15.0WEB
- https://github.com/mlflow/mlflow/security/advisories/GHSA-gqch-g4w5-7qcwWEB
- https://nvd.nist.gov/vuln/detail/CVE-2026-69148WEB
Advisories like this name the exposed products attackers go after on your perimeter. See what is exposed on your perimeter with a free passive scan.
Free scan