All vulnerabilities

Servers and applications

GHSA-gqvg-gmmx-x4hm

MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False safety control bypassed by mlflow.statsmodels flavor — RCE via crafted model artifact

Severity
High 8.8
Published
27 Jul 2026
Last changed
27 Jul 2026
Added to UpHeron
3 Nov 2026
Fixed in
3.15.0
Weakness
Deserialization of untrusted data (CWE-502)
Attack
Over the network · No login needed · Needs someone to click or open something
Attacker can
Read data · Change data · Take it down
CVSS vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
Sources
Project advisories (GitHub)

Description

MLflow introduced MLFLOW_ALLOW_PICKLE_DESERIALIZATION as a security control to prevent unsafe pickle.load execution during model loading, in response to CVE-2024-37052 through CVE-2024-37060. When set to False, operators expect all pickle deserialization to be blocked. The most recent related fix (#21188) patched a bypass in the pyfunc flavor.

However, the mlflow.statsmodels flavor completely omits this guard. An attacker who places a crafted MLmodel artifact into any accessible artifact store can trigger arbitrary code execution on any process that calls mlflow.pyfunc.load_model() against the malicious model — even when MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False.

This is a security control bypass. The operator believes pickle RCE is mitigated; the statsmodels flavor silently ignores the control.

mlflow.pyfunc.load_model() dispatches to flavor _load_pyfunc implementations via:

The guarded pattern (from mlflow/sklearn/init.py L526-533, the reference implementation) is:

From mlflow/mlflow's security advisory. Shortened; the full text is there.

Affected software

SoftwareAffected versions
MLflowServers and applications · github:mlflow/mlflow
  • From 2.1.0 before 3.15.0

About MLflow

Python library

From Wikidata, CC0 1.0.
Monthly downloads
30M+MLflow
Weekly downloads (PyPI)
5,332,947PyPI Stats, 15 Nov 2026
GitHub stars
28,816GitHub, 15 Nov 2026
Forks
6,611GitHub, 15 Nov 2026
Latest release
v3.18.0GitHub, 15 Nov 2026
Released
4 Nov 2026GitHub
License
Apache Software License 2.0Wikidata
First published on GitHub
5 Jun 2018GitHub

References