Vulnerability details
CVE-2026-5817
Docker Desktop
Docker Model Runner container-to-host code execution via unsandboxed trust_remote_code in Python inference backends
CVSS 4.0
8.8High
Official 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.
CVSS scores
| Score | Version | Source | Vector |
|---|---|---|---|
| 8.8High | 4.0 | CVEProject cvelistV5 (CNA) (Docker) | 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 |
| 8.2High | 3.1 | CVEProject cvelistV5 (CNA) (Docker) | CVSS:3.1/AV:L/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H |
Affected products
| Vendor | Product | Versions | Source |
|---|---|---|---|
| Docker | Docker Desktop | 4.62.0 ≤ v < 4.68.0 | CVEProject cvelistV5 (CNA) |
References and advisories
1- https://docs.docker.com/desktop/release-notes/#4680 release-notes
Kolun on Telegram @kolunsec
Exploited CVEs, key news and a daily digest
Alerts for this product's next CVEsFollow the product to get new vulnerabilities on Telegram.
Details
- Published
- 2026-05-22
- Last updated
- 2026-05-27
- State
- Published
- Assigner (CNA)
- Docker
Exploit prediction (EPSS)
0.22%
Probability of exploitation in 30 days
13.1%
Percentile (among all CVEs)
SSVC assessment (CISA)
- Automatable
- No
- Exploitation
- None
- Technical impact
- Total
Weaknesses (CWE)
- CWE-829Inclusion of Functionality from Untrusted Control Sphere