Strix
62.3K

CVE-2026-64849

CRITICALCVSS 9.3/10Actively ExploitedEPSS 8.15%

Last modified

CVE-2026-64849 is a critical-severity vulnerability rated 9.3/10 on the CVSS scale. MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. Prior to 3.15.0, the unauthenticated POST /api/2.0/mlflow/webhooks/{id}/test endpoint calls _validate_webhook_url() in mlflow/utils/validation.py only for the original URL while mlflow/webhooks/delivery.py follows redirects and re-resolves the hostname without pinning the validated address, allowing attackers to reach internal or cloud metadata services and receive response_status and response_body. CISA has confirmed active exploitation in the wild. EPSS estimates a 8.15% chance of exploitation in the next 30 days.

Description

MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. Prior to 3.15.0, the unauthenticated POST /api/2.0/mlflow/webhooks/{id}/test endpoint calls _validate_webhook_url() in mlflow/utils/validation.py only for the original URL while mlflow/webhooks/delivery.py follows redirects and re-resolves the hostname without pinning the validated address, allowing attackers to reach internal or cloud metadata services and receive response_status and response_body. This issue is fixed in version 3.15.0.

Metrics

CVSS 3.1
9.3/10

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:L/A:N

EPSS Probability
8.15%

94.4th percentile

Probability of exploitation in the next 30 days. Learn more

Exploitation Status

This vulnerability is listed in CISA’s Known Exploited Vulnerabilities catalog, confirming active exploitation in the wild. Federal agencies must remediate by .

Weakness Enumeration

Affected Software

VendorProductVersions
LfprojectsMlflow< 3.15.0

References

Timeline

Published
Last Modified
Status
Analyzed

Frequently Asked Questions

What is CVE-2026-64849?
MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. Prior to 3.15.0, the unauthenticated POST /api/2.0/mlflow/webhooks/{id}/test endpoint calls _validate_webhook_url() in mlflow/utils/validation.py only for the original URL while mlflow/webhooks/delivery.py follows redirects and re-resolves the hostname without pinning the validated address, allowing attackers to reach internal or cloud metadata services and receive response_status and response_body. This issue is fixed in version 3.15.0.
How severe is CVE-2026-64849?
CVE-2026-64849 has a CVSS score of 9.3/10 (CRITICAL severity). The EPSS model estimates a 8.15% probability of exploitation in the next 30 days. This vulnerability is listed in CISA's Known Exploited Vulnerabilities catalog.
How do I fix CVE-2026-64849?
Check the vendor references and advisories linked above for patched versions and mitigation guidance. You can also run a Strix scan to test if your systems are affected.

How Strix Helps

Related CVEs from 2026

View all 2026CVEs →

Are you affected by CVE-2026-64849?

Run a free Strix scan to check your systems for this vulnerability.

Scan your code now

Source: NVD / NIST