8.8
High
CVE-2026-5817 The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trust_remote_code=True when…
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.
Affected products and versions
| Product | Affected versions | Fixed in |
|---|---|---|
| Docker Desktop 4.62 | 4.62.0 → < 4.68.0 | 4.68.0 |
References
- docs.docker.com docs.docker.com
Timeline
- Published
- Last modified
Other CVEs of Docker Desktop
The 22 CVEs of Docker Desktop →
CVE-2026-112917
Docker Desktop WSL2 AF_VSOCK isolation bypass via 32-bit socketcall
9.2
Critical
CVE-2026-114409
Docker Desktop logs can expose stored third-party credentials
5.7
Medium
CVE-2026-106581
Before 4.92.0, Docker Desktop for Windows did not verify the signature of a package supplied to Docker Desktop…
8.2
High
CVE-2026-17106
The tar extraction routines in moby/go-archive (Unpack, UnpackLayer, Untar/UntarUncompressed, and the ApplyLayer…
7.1
High