CVE-2026-5817
Docker Model Runner container-to-host code execution via unsandboxed trust_remote_code in Python inference backends
- State
- Published
- Severity
- 8.8 High
- Published
- May 22, 2026
- Updated
- May 27, 2026
- Reserved
- Apr 8, 2026
Description
The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trust_remote_code=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.from_pretrained() to import and execute arbitrary Python files included in any model pulled from an OCI registry, resulting in arbitrary code execution on the Docker host as the Docker Desktop user when inference is triggered.
Any container on the Docker network can trigger this by calling the model-runner.docker.internal API to pull a malicious model and request inference.
Scores
| CVSS | Score | Vector | Source |
|---|---|---|---|
| 4.0 | 8.8 High | CVSS:4.0/AV:L/AC:L/AT:P/PR:L/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H | Docker |
| 3.1 | 8.2 High | CVSS:3.1/AV:L/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H | Docker |
Weaknesses
- CWE-829 Inclusion of Functionality from Untrusted Control Sphere
Affected Products
| Product | Versions |
|---|---|
Docker DesktopDocker |
|
References
- https://docs.docker.com/desktop/release-notes/#4680release-notes
Every record here comes from the CVE Program’s CVE List. © 1999–2026 The MITRE Corporation. CVE is a registered trademark of The MITRE Corporation. The records are reproduced under the CVE Program terms of use.