{"id":74623,"date":"2025-12-02T02:45:54","date_gmt":"2025-12-01T23:15:54","guid":{"rendered":"https:\/\/afaghhosting.net\/blog\/cve-2025-66448-vllm-vulnerable-to-remote-code-execution-via-transformers_utils-get_config\/"},"modified":"2025-12-02T02:45:54","modified_gmt":"2025-12-01T23:15:54","slug":"cve-2025-66448-vllm-vulnerable-to-remote-code-execution-via-transformers_utils-get_config","status":"publish","type":"post","link":"https:\/\/afaghhosting.net\/blog\/cve-2025-66448-vllm-vulnerable-to-remote-code-execution-via-transformers_utils-get_config\/","title":{"rendered":"CVE-2025-66448 &#8211; vLLM vulnerable to remote code execution via transformers_utils\/get_config"},"content":{"rendered":"<p>CVE ID : CVE-2025-66448<\/p>\n<p>Published :  Dec. 1, 2025, 11:15 p.m. | 1\u00a0hour, 9\u00a0minutes ago<\/p>\n<p>Description : vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.11.1, vllm has a critical remote code execution vector in a config class named Nemotron_Nano_VL_Config. When vllm loads a model config that contains an auto_map entry, the config class resolves that mapping with get_class_from_dynamic_module(&#8230;) and immediately instantiates the returned class. This fetches and executes Python from the remote repository referenced in the auto_map string. Crucially, this happens even when the caller explicitly sets trust_remote_code=False in vllm.transformers_utils.config.get_config. In practice, an attacker can publish a benign-looking frontend repo whose config.json points via auto_map to a separate malicious backend repo; loading the frontend will silently run the backend\u2019s code on the victim host. This vulnerability is fixed in 0.11.1.<\/p>\n<p>Severity: 7.1 | HIGH<\/p>\n<p>Visit the link for more details, such as CVSS details, affected products, timeline, and more&#8230;\u00a0<\/p>\n","protected":false},"excerpt":{"rendered":"<p>CVE ID : CVE-2025-66448 Published : Dec. 1, 2025, 11:15 p.m. | 1\u00a0hour, 9\u00a0minutes ago Description : vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.11.1, vllm has a critical remote code execution vector in a config class named Nemotron_Nano_VL_Config. When vllm loads a model config that contains an &hellip;<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[26],"tags":[],"class_list":["post-74623","post","type-post","status-publish","format-standard","hentry","category-vulnerability"],"_links":{"self":[{"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/posts\/74623","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/comments?post=74623"}],"version-history":[{"count":0,"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/posts\/74623\/revisions"}],"wp:attachment":[{"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/media?parent=74623"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/categories?post=74623"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/tags?post=74623"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}