CVE 8.8 HIGH

Docker Model Runner container-to-host code execution via unsandboxed trust_remote_code in Python inference backends_CVE-2026-5817

8.8 / 10
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

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.

AI Analysis

Unsandboxed code execution vulnerability in Docker Model Runner

Basic Information

ID CVE-2026-5817
Source Docker
Published May 22, 2026 at 19:24

Affected Product

Vendor Docker
Product Docker Desktop
Version 4.62.0
Affected Versions Docker Docker Desktop 4.62.0

CWE Classification

AI Assessment

AI Score 8.8 / 10
AI Severity High
Vendor Docker
Product Docker Model Runner
Version 4.62.0

References

💭 Join the Security Discussion

🔒 Your email address will not be published. Required fields are marked *

⚠️ Please be respectful and constructive in your comments. Security discussions should remain professional.