Vulnerabilities · 2026 · Vllm
CVE-2026-22807
Vllm vulnerability
vLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1 and prior to version 0.14.0, vLLM loads Hugging Face `auto_map` dynamic modules during model resolution without gating on `trust_remote_code`, allowing attacker-controlled Python code in a model repo/path to execute at server startup. An attacker who can influence the model repo/path (local directory or remote Hugging Face repo) can achieve arbitrary code execution on the vLLM host during model load. This happens before any request handling and does not require API access. Version 0.14.0 fixes the issue.
Affected products
Vllm. The exact versions are in the references.
What to do now
- Check your exposureFind out whether Vllm runs anywhere in your environment, including systems of suppliers.
- UpdateInstall the version in which the vendor fixed the vulnerability. The references lead to the advisory.
- Limit access until thenIf you cannot update at once, take the system off the internet or restrict who can reach it.
References
- github.com/vllm-project/vllm/security/advisories/GHSA-2pc9-4j83-qjmr · Patch
- github.com/vllm-project/vllm/commit/78d13ea9de4b1ce5e4d8a5af9738fea71fb024e5 · Patch
- github.com/vllm-project/vllm/pull/32194 · Patch
- github.com/vllm-project/vllm/releases/tag/v0.14.0
- nvd.nist.gov/vuln/detail/CVE-2026-22807
- github.com/advisories/GHSA-2pc9-4j83-qjmr
- access.redhat.com/errata/RHSA-2026:10184
- access.redhat.com/errata/RHSA-2026:30087
- nvd.nist.gov/vuln/detail/CVE-2026-22807
- cve.org/CVERecord?id=CVE-2026-22807
More vulnerabilities in Vllm products
Data: CVE list and the NVD. This product uses data from the NVD API but is not endorsed or certified by the NVD.