Vulnerability 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 Score
HIGH
Affected Products
| Vendor | Product | Versions |
|---|---|---|
| Vllm | Vllm | < 0.28.0 |
Related Weaknesses (CWE)
References
- https://github.com/vllm-project/vllm/security/advisories/GHSA-3c86-2m5g-59q7Vendor Advisory
- https://www.vulncheck.com/advisories/vllm-before-0.28.0-remote-code-execution-viThird Party Advisory
FAQ
What is CVE-2026-90553?
CVE-2026-90553 is a vulnerability with a CVSS score of 7.8 (HIGH). 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 ...
How severe is CVE-2026-90553?
CVE-2026-90553 has been rated HIGH with a CVSS base score of 7.8/10. Review the CVSS metrics above for detailed severity breakdown.
Is there a patch for CVE-2026-90553?
Check the references section above for vendor advisories and patch information. Affected products include: Vllm Vllm.