CVE-2026-12491
vLLM: image EXIF Rotation & PNG tRNS Transparency Not Normalized, Causing Mismatch Between Model Input and Expectations
Auswirkung
A flaw was found in vLLM, an open-source library for large language model inference. This vulnerability arises from improper handling of image metadata, specifically EXIF orientation and PNG transparency (tRNS) data, during image processing. When images are converted to RGB, transparency information may be implicitly discarded or remapped, leading to unexpected rendering of transparent pixels and distortion of input content. This can result in the model misinterpreting image content, potentially affecting the integrity of processed data.
Ein Angreifer kann die Lücke über das Netzwerk ausnutzen, ohne Anmeldung, ohne Zutun eines Nutzers. Dafür müssen bestimmte Bedingungen erfüllt sein, was einen Angriff erschwert. Bei Erfolg: geringes Risiko, dass Daten oder Systeme verändert werden, geringes Risiko für Ausfälle.
Betroffene Pakete
| Ökosystem | Paket | Betroffen | Behoben in |
|---|---|---|---|
| pip | vllm | >= 0.11.0, < 0.24.0 | 0.24.0 |
Schwachstellenklasse
Referenzen
- NVD-Eintrag
- CVE.org-Eintrag
- GitHub-Advisory
- https://github.com/advisories/GHSA-8jr5-v98p-w75m
- https://github.com/vllm-project/vllm/security/advisories/GHSA-8jr5-v98p-w75m
- https://github.com/vllm-project/vllm/pull/44974
- https://github.com/vllm-project/vllm/commit/cf1c90672404548aa3bc51f92c4745576a65ee26
- https://nvd.nist.gov/vuln/detail/CVE-2026-12491
- https://access.redhat.com/security/cve/CVE-2026-12491
- https://bugzilla.redhat.com/show_bug.cgi?id=2489786
- https://github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2026-3406.yaml
- https://github.com/vllm-project/vllm
- https://pypi.org/project/vllm