Skip to main content
← Vulnerabilities

MLflow PickleEvaluationArtifact Deserialization of Untrusted Data Remote Code Execution Vulnerability

Fixed in MLflow 3.13.0. No vendor severity rating is published for this vulnerability.

Written by anthropic/claude-haiku-5.5 from the sources below. Verify against them.

7.8HighCNA
Exploited
No known exploitation
Fix
Fixed in 3.13.0
Who is exposed
Specific setups only

Do first: Upgrade to a fixed version: 3.13.0.

13Pcve.now Index 13 of 100 · methodology v3 · provisional
  • Exploitation?/40Exploitation: unknown (data gap). No CISA SSVC exploitation rating yet
  • Exposure5/25Specific setups only
  • Reach?/20Reach: unknown (data gap). No tracked technology or package matched
  • Severity7.8/10CVSS 7.8
  • No patch0/5No signal

What would change the score

  • Would score about 63 if added to CISA KEV
  • Would score about 58 if CISA's SSVC rates exploitation active
  • Would score about 23 if CISA's SSVC rates it automatable

CVE-2026-118623 is not in CISA's Known Exploited Vulnerabilities (KEV) catalog. No public exploit code is tracked. It is rated CVSS 7.8 (high) and affects MLflow; fixed in 3.13.0.

Published
Last modified
Published
Not listed in CISA KEV
13/100 · provisional
Specific setups only
Fix
Fixed in 3.13.0
None on Exploit-DB, Metasploit or GitHub

Summary

MLflow, an open-source machine learning platform, is affected by this vulnerability. An attacker could exploit it by tricking a user into visiting a malicious page or opening a malicious file, allowing arbitrary code to run within the context of the MLflow process.

Read the original description

MLflow PickleEvaluationArtifact Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of MLflow. User interaction is required to exploit this vulnerability in that the target must visit a malicious page or open a malicious file. The specific flaw exists within the parsing of pickle files. The issue results from the lack of proper validation of user-supplied data, which can result in deserialization of untrusted data. An attacker can leverage this vulnerability to execute code in the context of the current process. Was ZDI-CAN-29513.

What to do

  • Upgrade to a fixed version: 3.13.0.

Timeline

  1. CVE published
  2. First seen by cve.now

Affected versions

  • MLflow MLflow

    3.10.1

Evidence 1

  1. Link

Corrections

None so far.

Something wrong on this page? Report an error. Every fix is listed here.

FAQ

Questions about CVE-2026-118623

Is CVE-2026-118623 being exploited?

It is not in CISA's Known Exploited Vulnerabilities catalog, and CISA's SSVC assessment does not mark exploitation as active. That does not mean it has not been attacked: vendors and researchers can report attacks before CISA lists them. No public exploit code is tracked.Source: cve.now feed data

What does this mean?

Loading untrusted serialized data through PickleEvaluationArtifact can run attacker-supplied code inside the current MLflow process. The advisory describes this as code execution in the context of the running process.Source: zerodayinitiative.com

Which versions are affected by CVE-2026-118623?

MLflow: version 3.10.1.Source: cve.now feed data

Am I affected, and how do I check?

The checked facts do not list affected version ranges or specific setups. To check an installation, run `mlflow --version` or `pip show mlflow` and compare the result with the fixed version, 3.13.0.Source: zerodayinitiative.com

How do I fix CVE-2026-118623?

Upgrade MLflow to 3.13.0 or later, which contains the fix.Source: zerodayinitiative.com