vLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1 and prior to version 0.18.0, two model implementation files hardcode `trust_remote_code=True` when loading sub-components, bypassing the user's explicit `--trust-remote-code=False` security opt-out. This enables remote code execution via malicious model repositories even when the user has explicitly disabled remote code trust. Version 0.18.0 patches the issue.
01
beschreibt (1)
Zugehörige CVEs.
| ID | Schweregrad | ausnutzbar | ||
|---|---|---|---|---|
| CVE-2026-27893US | — | — | ||
vLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1 and prior to version 0.18.0, two model implementation files hardcode `trust_rem… | ||||
02
Referenzen
Externe Belege — Patches, Meldungen, Analysen.
- https://access.redhat.com/errata/RHSA-2026:10140ADVISORY
- https://access.redhat.com/errata/RHSA-2026:10141ADVISORY
- https://access.redhat.com/errata/RHSA-2026:19712ADVISORY
- https://access.redhat.com/errata/RHSA-2026:19724ADVISORY
- https://access.redhat.com/errata/RHSA-2026:19725ADVISORY
- https://access.redhat.com/errata/RHSA-2026:24977ADVISORY
- https://access.redhat.com/errata/RHSA-2026:37275ADVISORY
- https://access.redhat.com/errata/RHSA-2026:8746ADVISORY
scry
Advisories wie diese benennen die exponierten Produkte, die Angreifer auf Ihrer Angriffsfläche ins Visier nehmen. Sehen Sie mit einem kostenlosen Passiv-Scan, was exponiert ist.
Kostenloser Scan