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.
No advisories yet.
Solution
No solution given by the vendor.
Workaround
Disable Docker Model Runner or only run trusted containers on Docker Desktop instances where Model Runner is enabled.
| Link | Providers |
|---|---|
| https://docs.docker.com/desktop/release-notes/#4680 |
|
Tue, 26 May 2026 16:30:00 +0000
| Type | Values Removed | Values Added |
|---|---|---|
| Metrics |
ssvc
|
Fri, 22 May 2026 21:45:00 +0000
| Type | Values Removed | Values Added |
|---|---|---|
| First Time appeared |
Docker
Docker docker Desktop |
|
| Vendors & Products |
Docker
Docker docker Desktop |
Fri, 22 May 2026 20:00:00 +0000
| Type | Values Removed | Values Added |
|---|---|---|
| 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. | |
| Title | Docker Model Runner container-to-host code execution via unsandboxed trust_remote_code in Python inference backends | |
| Weaknesses | CWE-829 | |
| References |
| |
| Metrics |
cvssV3_1
|
Projects
Sign in to view the affected projects.
Status: PUBLISHED
Assigner: Docker
Published:
Updated: 2026-05-27T03:55:37.087Z
Reserved: 2026-04-08T15:34:05.200Z
Link: CVE-2026-5817
Updated: 2026-05-26T15:21:52.749Z
Status : Awaiting Analysis
Published: 2026-05-22T20:16:35.120
Modified: 2026-05-22T20:44:05.750
Link: CVE-2026-5817
No data.
OpenCVE Enrichment
Updated: 2026-05-22T22:00:12Z