CVE-2026-73560: vLLM is an inference and serving engine for large language models

Routine

MEDIUM 6.5UPGRADE

CVE-2026-73560 is a medium-severity flaw (CVSS 6.5) affecting vllm. Below: why it ranks where it does, which versions are affected, what to do, and the exact commands.

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the MiMoV2OmniMultiModalProcessor in vllm/transformers_utils/processors/mimo_v2_omni.py passes attacker-controlled image and audio strings through _fetch_image, requests.get, and Image.open instead of MediaConnector, bypassing allowed_media_domains and allowed_local_media_path protections and allowing server-side requests and reads of arbitrary files accessible to

Why this priority

CISA SSVC: public proof of concept

A fixed release or distro patch exists.

Risk facts

CVSS score	6.5 (Medium)
Attack vector	Network (remote)
Attack complexity	Low
Privileges needed	Low
User interaction	None
EPSS exploit chance	0.4% (top 66%)
CISA exploited list	No
Weakness	CWE-918
Published	2026-08-17
Last modified	2026-10-23

Publisher (CNA): GitHub, Inc. · Vendor: vllm

Last updated 2026-10-23 · Sources: CVE Program, NVD, CISA KEV, EPSS and vendor advisories · How we score

Am I affected?

Timeline

2026-08-17 21:16 UTCPublished CVE record published
2026-10-23 16:10 UTCUpdated Record last modified

vllm vllm UPGRADE

Affected: < 0.26.0

Fixed in: 0.26.0

Safe to downgrade to: no unaffected older release

1. Check installed version

apt-cache policy vllm || rpm -q vllm || vllm --version

2. UPGRADE to 0.26.0 (Debian / Ubuntu)

sudo apt update && sudo apt install --only-upgrade vllm

2. UPGRADE (RHEL / Amazon Linux / Fedora)

sudo dnf upgrade vllm

Note

Distros often backport fixes without changing the version number. Confirm with the distro advisory or changelog.

Exact fixes by distribution

Python (pip) · vllm · FIXED · 0.26.0 · vulnerable: < 0.26.0 · source

pip install --upgrade "vllm>=0.26.0" pip show vllm

Fedora 46 · gum · FIXED · 2.0.2-1.fc46 · FEDORA-2026-da0d6dd894 · source

sudo dnf upgrade --advisory=FEDORA-2026-da0d6dd894

RHEL AI (RHEL AI) 3 · rhelai3/bootc-aws-cuda-rhel9 · DEFERRED

RHEL AI (RHEL AI) 3 · rhelai3/bootc-azure-cuda-rhel9 · DEFERRED

RHEL AI (RHEL AI) 3 · rhelai3/bootc-azure-rocm-rhel9 · DEFERRED

RHEL AI (RHEL AI) 3 · rhelai3/bootc-cuda-rhel9 · DEFERRED

RHEL AI (RHEL AI) 3 · rhelai3/bootc-gaudi-rhel9 · DEFERRED

Verify at the official source NVD · CVE Program · Report a problem with this page
Scores and fixes here are computed automatically from those and other public sources. Confirm with the vendor advisory before changing production systems.

Quick answers

How do I fix CVE-2026-73560?

vllm: upgrade to 0.26.0. Exact commands for each OS are on this page.

Is CVE-2026-73560 being exploited?

CVE-2026-73560 is not on the CISA Known Exploited Vulnerabilities list at the time of the last update. Check again, as this can change.

What does CVE-2026-73560 affect?

CVE-2026-73560 affects vllm. Severity is medium (CVSS 6.5). vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the MiMoV2OmniMultiModalProcessor in vllm/transformers_utils/processors/mimo_v2_omni.py passes attacker-controlled image and audio strings through _fetch_im

How urgent is CVE-2026-73560?

Priority 34/100. CISA SSVC: public proof of concept

Related CVEs

CVE-2026-56340 HIGH 8.8 UPGRADE vLLM versions >= 0.10.2 and < 0.13.0 are missing sparse tensor validation in multimodal embeddings processing
CVE-2026-71486 MEDIUM 4.3 UPGRADE vLLM is an inference and serving engine for large language models

References

← Back to latest CVEs