CVE-2026-116692 MLflow: Remote code execution on unauthenticated MLflow servers via third-party scorer deserialization importing caller-placed artifact modules
- Published
- Updated
- Exploit status
- No public exploit known
Description
MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. From 3.12.0rc0 until 3.17.0, third-party scorer deserialization in mlflow/genai/scorers/base.py accepts request-selected descendants of THIRD_PARTY_SCORER_ALLOWED_MODULES rather than requiring an exact allowed module. A caller can use MLflow artifact endpoints to place a Python module beneath a writable allowed package and submit its module and class through /ajax-api/3.0/mlflow/scorer/invoke. Scorer.model_validate_json() imports the module and constructs the class before trace processing. Authentication-disabled deployments permit this code execution without credentials, while basic-auth deployments require a principal capable of creating an experiment and run and uploading artifacts. In either configuration, the allowed scorer package must be writable by the server process. This issue is fixed in version 3.17.0.
Affected products
| Product | Affected versions | Fixed in |
|---|---|---|
| pypi mlflow(PyPI) | >= 3.12.0rc0, < 3.17.0 | 3.17.0 |
| pypi mlflow(PyPI) | = 3.12.0 | No fix listed |
| pypi mlflow(PyPI) | = 3.12.0rc0 | No fix listed |
| pypi mlflow(PyPI) | = 3.13.0 | No fix listed |
| pypi mlflow(PyPI) | = 3.13.0rc0 | No fix listed |
| pypi mlflow(PyPI) | = 3.14.0 | No fix listed |
| pypi mlflow(PyPI) | = 3.15.0 | No fix listed |
| pypi mlflow(PyPI) | = 3.15.1 | No fix listed |
| pypi mlflow(PyPI) | = 3.15.2 | No fix listed |
| pypi mlflow(PyPI) | = 3.16.0 | No fix listed |
| pypi mlflow(PyPI) | = 3.16.1 | No fix listed |
| redhat/rhoai/odh-mlflow-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-pipeline-runtime-datascience-cpu-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-pipeline-runtime-pytorch-cuda-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-pipeline-runtime-pytorch-llmcompressor-cuda-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-pipeline-runtime-pytorch-rocm-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-pipeline-runtime-tensorflow-cuda-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-pipeline-runtime-tensorflow-rocm-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-th-torch-cpu-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-th-torch-cuda-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-th-torch-rocm-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-workbench-codeserver-datascience-cpu-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-workbench-jupyter-datascience-cpu-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-workbench-jupyter-pytorch-cuda-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-workbench-jupyter-pytorch-llmcompressor-cuda-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
Show 5 more
| Product | Affected versions | Fixed in |
|---|---|---|
| redhat/rhoai/odh-workbench-jupyter-pytorch-rocm-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-workbench-jupyter-tensorflow-cuda-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-workbench-jupyter-tensorflow-rocm-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| redhat/rhoai/odh-workbench-jupyter-trustyai-cpu-py312-rhel9(Red Hat:openshift_ai) | All versions | No fix listed |
| mlflow | = >= 3.12.0rc0, < 3.17.0 | No fix listed |
References
- https://github.com/mlflow/mlflow/security/advisories/GHSA-26p8-2jq9-3vq9 vendor-advisory
- https://github.com/mlflow/mlflow/pull/25906 patch
- https://github.com/mlflow/mlflow/commit/a732443e695c3c677680c54320efc394c81654ab patch
- github.com product
- https://github.com/mlflow/mlflow/releases/tag/v3.17.0
- Red Hat CVE page mitigation vendor-advisory
- Red Hat Bugzilla 2562393 issue-tracking
- www.cve.org
- nvd.nist.gov
- github.com