8.5

CVE-2026-90553: vLLM before 0.28.0 Remote Code Execution via LlavaOnevision2 processor

Analyzed
AliasesPYSEC-2026-3985GHSA-3c86-2m5g-59q7
Published 12-09-2026Last modified 14-09-2026

Vulnerability summary

CVE-2026-90553 is a high (CVSS 8.5) Code Injection vulnerability in vllm-project vllm. EPSS estimates a 0.31% chance of exploitation in the next 30 days, higher than 22.4% of scored vulnerabilities.

Overall risk
34/100
0100
Medium · default modelHow is this calculated?
CVSS score
8.5/10
010
High · v4.0 (CVE.org)
EPSS score
0.31%
0%100%
Low · +0.10 pp in 30 days
KEVNot listedNot listed by any tracked catalog.
RansomwareNot flaggedNo ransomware use recorded.
Public exploits0No public exploits recorded

Timeline

  1. 12-09-2026
    PublishedVulnerability first disclosed.
  2. 14-09-2026
    Last modifiedVulnerability information updated.

Description

vLLM before 0.28.0 contains a remote code execution vulnerability in the LlavaOnevision2 processor loader that ignores the trust_remote_code parameter when loading remote processor classes. Attackers can craft a malicious model with arbitrary code in processing_llava_onevision2.py that executes with vLLM process authority even when trust_remote_code is set to False.

CVSS metrics

5 scores · 3 sources · 2 versions

EPSS trend

Probability

0.31%

Chance of exploitation in the next 30 days

Percentile

22%

Versus all other CVEs

30-day change

Increase +0.10 pp

From 0.21% on 22-09-2026

22-09-202622-10-2026

Known exploited

Compare KEV catalogs
CISA KEV

Not in the CISA KEV catalog.

ENISA CNW

Not in the ENISA CNW catalog.

CIRCL

Not in the CIRCL catalog.

Public exploits

0 found

No public exploits indexed.

Weaknesses

CWE · root cause

Affected systems

PyPI1 product
  • vllm< 0.28.0
Vllm1 product
Vllm-project1 product

3 vendors · 3 products

References

2 links