CVE-2026-22807 | vLLM affected by RCE via auto_map dynamic module loading during model initialization

vLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1 and prior to version 0.14.0, vLLM loads Hugging Face `auto_map` dynamic modules during model resolution without gating on `trust_remote_code`, allowing attacker-controlled Python code in a model repo/path to execute at server startup. An attacker who can influence the model repo/path (local directory or remote Hugging Face repo) can achieve arbitrary code execution on the vLLM host during model load. This happens before any request handling and does not require API access. Version 0.14.0 fixes the issue.

Published: 2026-01-21 Last update: 2026-01-30 Assigner: [email protected] Source: [email protected]

Conclusion & alert: CVE-2026-22807 is rated Low Risk (37/100): CVSS High severity, with low exploitation likelihood (EPSS 0.02%). Mandatory action: Monitor for updates and reassess as exploit intelligence or EPSS changes.

Risk is dynamic; we continuously reassess and refresh what is shown on this page as upstream context changes.

Exploit prediction scoring system (EPSS) score for CVE-2026-22807

EPSS lead: Daily EPSS estimates relative likelihood of exploitation; percentile ranks this CVE among scored vulnerabilities (higher = more severe relative rank).

# Date Old EPSS score New EPSS score Delta (New - Old)
1 2026-03-21 0.05% 0.02% -0.03%
2 2026-01-27 0.05% 0.05% +0.00%
3 2026-01-22 0.05%

Full EPSS history (3 records total)

Common vulnerability scoring system (CVSS) metrics for CVE-2026-22807

CVSS metrics for this CVE.

Base score Version Severity Vector Exploitability Impact Score source
8.8 3.1 HIGH
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H Click to expand
Attack vector (AV:N)
Could be attacked over the internet or any normal routed network—not just someone sitting at the machine.
Attack complexity (AC:L)
Once they can reach the bug, pulling it off is straightforward—no weird race conditions or rare setup.
Privileges required (PR:N)
No account or special rights needed—anonymous or random user is enough.
User interaction (UI:R)
A real person has to do something—click, install, enable—otherwise it doesn’t land.
Scope (S:U)
Damage stays in the same “trust bubble” as the broken component—no big spill into unrelated systems.
Confidentiality (C:H)
Serious risk that confidential data gets exposed in a big way.
Integrity (I:H)
They could widely tamper with or forge data—trust in the data is badly hurt.
Availability (A:H)
Could take the service down hard or make it unusable for people who depend on it.
2.8 5.9 [email protected]
9.8 3.1 CRITICAL
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H Click to expand
Attack vector (AV:N)
Could be attacked over the internet or any normal routed network—not just someone sitting at the machine.
Attack complexity (AC:L)
Once they can reach the bug, pulling it off is straightforward—no weird race conditions or rare setup.
Privileges required (PR:N)
No account or special rights needed—anonymous or random user is enough.
User interaction (UI:N)
Nobody has to click “OK” or open a trap file; it can work without a victim helping.
Scope (S:U)
Damage stays in the same “trust bubble” as the broken component—no big spill into unrelated systems.
Confidentiality (C:H)
Serious risk that confidential data gets exposed in a big way.
Integrity (I:H)
They could widely tamper with or forge data—trust in the data is badly hurt.
Availability (A:H)
Could take the service down hard or make it unusable for people who depend on it.
3.9 5.9 [email protected]

Weakness enumeration for CVE-2026-22807

GitHub Security Advisory for CVE-2026-22807

GHSA-2pc9-4j83-qjmr · Severity: high · Ecosystem: pip — vLLM affected by RCE via auto_map dynamic module loading during model initialization

OS Trackers for CVE-2026-22807

vendor priority summary link
redhat high https://access.redhat.com/security/cve/CVE-2026-22807

Affected software / configurations for CVE-2026-22807

Vendor Product Version Raw CPE
vllm vllm >= 0.10.1, < 0.14.0 cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*

References for CVE-2026-22807

cvelogic Threat Intelligence