CVE-2026-85673
Received Received - Intake

SSRF Bypass in LLaMA-Factory via Redirect Handling

Publication date: 2026-09-04

Last updated on: 2026-09-04

Assigner: VulnCheck

Description
LLaMA-Factory contains a server-side request forgery vulnerability in the OpenAI-compatible API multimodal media URL handler that allows unauthenticated attackers to bypass SSRF validation. The check_ssrf_url guard validates URLs once but requests.get follows redirects and re-resolves DNS without re-validation, enabling attackers to use HTTP redirects or DNS rebinding to access internal addresses and cloud metadata endpoints.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published
2026-09-04
Last Modified
2026-09-04
Generated
2026-09-04
AI Q&A
2026-09-04
EPSS Evaluated
N/A
NVD
CVE-2026-85673
EUVD
EUVD-2026-71129

Affected Vendors & Products

Showing 2 associated CPEs
Vendor Product Version / Range
hiyouga llamafactory 0.9.5
hiyouga llamafactory to 0.9.5 (exc)

Helpful Resources

Exploitability

CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-918 The web server receives a URL or similar request from an upstream component and retrieves the contents of this URL, but it does not sufficiently ensure that the request is being sent to the expected destination.

Attack-Flow Graph

AI Quick Actions

Instant insights powered by AI
Executive Summary

CVE-2026-85673 is a Server-Side Request Forgery (SSRF) vulnerability in LLaMA-Factory's OpenAI-compatible API multimodal media URL handler. The SSRF validation check is performed only once, but the subsequent requests.get call follows HTTP redirects and re-resolves DNS without re-validation. This allows unauthenticated attackers to bypass SSRF protections using HTTP redirects or DNS rebinding to access internal network addresses and cloud metadata endpoints.

Detection Guidance

To detect this SSRF vulnerability, monitor network traffic for unusual outbound requests from the LLaMA-Factory API server, especially to internal or cloud metadata endpoints. Check logs for HTTP redirect chains or DNS rebinding attempts targeting the API's multimodal media URL handler. Use tools like tcpdump or Wireshark to capture traffic from the API process.

Impact Analysis

This vulnerability allows attackers to access internal network resources and cloud metadata endpoints without authentication. It could lead to data breaches, unauthorized information disclosure, or further exploitation of internal systems. The high CVSS score (8.7) indicates significant risk of exploitation and potential impact on confidentiality.

Compliance Impact

This vulnerability could lead to unauthorized access to sensitive data, violating GDPR's data protection requirements and HIPAA's security rules for protected health information. Organizations using affected versions may face compliance violations, regulatory penalties, and loss of customer trust due to potential data exposure.

Mitigation Strategies

Immediately upgrade LLaMA-Factory to a patched version if available. If not, disable the OpenAI-compatible API multimodal media URL handler or restrict network access to the API server. Implement additional SSRF protections like validating URLs after redirects and blocking private/reserved IP access in the firewall.

Chat Assistant

Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-85673. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70

EPSS Chart