Machine-written summary
Summary written by Claude Haiku from the source advisory. Not reviewed by a person.
vLLM versions before 0.11.1 are vulnerable to remote code execution through the Nemotron_Nano_VL_Config config class. When a model config contains an auto_map entry, the class fetches and runs Python code from the repository that entry names, even with trust_remote_code set to False. An attacker who publishes a model repo that points to a malicious second repo can run code on the host of anyone who loads it. Upgrade to vLLM 0.11.1 or later.
Not analysed yet
We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.
Severity and exploitation
| Score | Rating | CVSS | Vector | Source |
|---|---|---|---|---|
| 7.1 | High | 3.1 | CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:U/C:H/I:H/A:H | CNA (GitHub_M) |
- CISA KEV Not listed.
- EPSS 0.0071, the 54th percentile, on 15 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.
Timeline
- 31 Oct 2026, 14:42 UTC First seen by agentprime
- 31 Oct 2026, 14:42 UTC Confirmed by OSV
Sources
- GitHub Advisory Database: GHSA-8fr4-5q9j-m8gm
- OSV: GHSA-8fr4-5q9j-m8gm
- github.com/vllm-project/vllm/security/advisories/GHSA-8fr4-5q9j-m8gm
- nvd.nist.gov/vuln/detail/CVE-2025-66448
- github.com/vllm-project/vllm/pull/28126
- github.com/vllm-project/vllm/commit/ffb08379d8870a1a81ba82b72797f196838d0c86
- github.com/advisories/GHSA-8fr4-5q9j-m8gm
- github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2026-2015.yaml
- github.com/vllm-project/vllm
- pypi.org/project/vllm