CVE-2026-25960 is a high-severity (CVSS 7.1) Server-Side Request Forgery (SSRF) vulnerability in vllm. A fix is available for vllm — see the affected versions and patch details below.
SSRF Protection Bypass in vLLM
Exploitation Status
No confirmed exploitation observed yet
- CISA’s own triage has not observed active exploitation or public proof-of-concept code for this CVE as of its last assessment.
Exploitation and automatability from CISA’s SSVC triage for CVE-2026-25960.
EPSS Exploitation Probability
Probability of exploitation in the next 30 days, from FIRST.org EPSS.
How urgent is this, really
Where this sits among everything scored
Real-World Exposure
vllmReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects PyPI packages — download data is not available via public APIs for these ecosystems.
Description
Summary
The SSRF protection fix for https://github.com/vllm-project/vllm/security/advisories/GHSA-qh4c-xf7m-gxfc can be bypassed in the load_from_url_async method due to inconsistent URL parsing behavior between the validation layer and the actual HTTP client.
Affected Component
- File:
vllm/connections.py - Function:
load_from_url_async
Vulnerability Details
Root Cause
The SSRF fix uses urllib3.util.parse_url() to validate and extract the hostname from user-provided URLs. However, load_from_url_async uses aiohttp for making the actual HTTP requests, and aiohttp internally uses the yarl library for URL parsing.
These two URL parsers handle backslash characters (\) differently:
| Parser | Input URL | Parsed Host | Parsed Path | Behavior |
|---|---|---|---|---|
urllib3.parse_url() | https://httpbin.org\@evil.com/ | httpbin.org | /%[email protected]/ | URL-encodes \ as %5C, treats \@evil.com/ as part of the path |
yarl (via aiohttp) | https://httpbin.org\@evil.com/ | evil.com | / | Treats \ as part of userinfo (user: httpbin.org\), the @ acts as the userinfo/host separator |
Attack Scenario
# Attacker provides this URL
malicious_url = "https://httpbin.org\\@evil.com/"
# 1. Validation layer (urllib3.parse_url)
parsed = urllib3.util.parse_url(malicious_url)
# parsed.host == "httpbin.org" ✅ Passes validation
# 2. Actual request (aiohttp with yarl)
async with aiohttp.ClientSession() as session:
async with session.get(malicious_url) as response:
# Request actually goes to evil.com! ❌ Bypass!
Why This Happens
- yarl: Interprets
httpbin.org\as the userinfo component, and@as the userinfo/host separator, so the URL is parsed asuser=httpbin.org\,host=evil.com,path=/ - urllib3: URL-encodes the backslash as
%5C, so\@evil.com/becomes/%[email protected]/which is treated as part of the path, leavinghost=httpbin.org
This inconsistency allows an attacker to:
- Bypass the hostname allowlist check
- Access arbitrary internal/external services
- Perform full SSRF attacks
Fixes
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐍PyPI | vllm | ≥ 0.15.1&&< 0.17.0 | 0.17.0pip install --upgrade 'vllm==0.17.0' |
Affected Products
vllmvllmDetection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for vllm, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.
Fix
Update vllm to 0.17.0 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms CVE-2026-25960 is resolved across your whole dependency graph.
Workarounds
Restrict outbound requests from the affected component to an allowlist of hosts, block access to link-local and internal address ranges at the network layer, and require authentication on internal services so a forged request cannot reach them unauthenticated.
Fixing This On Your OS
If you run this on a Linux distribution, patch through your package manager against the distro's own security advisory below — it tracks the exact backported fix for your release, which can ship on a different timeline (and sometimes a different severity) than the upstream project.
This is an IMPORTANT vulnerability. The vLLM component, as used in Red Hat AI Inference Server and Red Hat OpenShift AI, is affected by a Server-Side Request Forgery (SSRF) protection bypass. This flaw allows an attacker to circumvent existing SSRF safeguards due to inconsistent URL parsing between the validation…
Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability.Source: Red Hat security advisory for CVE-2026-25960 (CC BY 4.0)
| Product | Fixed in | Advisory |
|---|---|---|
| Red Hat OpenShift AI 2.25 | rhoai/odh-vllm-gaudi-rhel9:1780069069 | RHSA-2026:24977 |
| Red Hat OpenShift AI 2.25 | rhoai/odh-vllm-cuda-rhel9:1783998774 | RHSA-2026:42644 |
| Red Hat OpenShift AI 2.25 | rhoai/odh-vllm-cpu-rhel9:1792018436 | RHSA-2026:87410 |
Frequently Asked Questions
Is CVE-2026-25960 in your dependencies?
Find it across PyPI, including transitive dependencies.