THREATOPS
THREAT OPS › Threat News › Hostile Inference: Exploiting Docker Model Runner for Container Escapes

Hostile Inference: Exploiting Docker Model Runner for Container Escapes

medzdi_blogPublished 2026-11-18

<p class="">Local LLM inference is seeing adoption in default developer workflows, and the tooling that serves those models now ships inside products developers already trust. One of those is Docker Model Runner (DMR), the feature in Docker Desktop that pulls, runs, and serves models locally. In November 2025, we reported <a href="https://www.zerodayinitiative.com/advisories/ZDI-26-150/">ZDI-26-15

MITRE ATT&CK techniques

Indicators of compromise

Original source: https://www.thezdi.com/blog/2026/11/17/hostile-inference-exploiting-docker-model-runner-for-container-escapes