CVE-2025-66448
CVE-2025-66448 is a high-severity vulnerability in Vllm with a CVSS 3.x base score of 7.1. It is not currently listed as actively exploited by CISA, and its EPSS exploit-prediction score is low. The underlying weakness is classified as CWE-94.
Key facts
- Severity: High (CVSS 3.x base score 7.1)
- EPSS exploit prediction: 1% (44th percentile)
- Actively exploited: Not listed in CISA KEV
- EU (EUVD) id: EUVD-2025-200115
- Weakness: CWE-94
- Affected product: Vllm
- Published:
- Last modified:
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(...) 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’s code on the victim host. This vulnerability is fixed in 0.11.1.
Frequently asked questions
- What is CVE-2025-66448?
- 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(...) 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’s code on the victim host. This vulnerability is fixed in 0.11.1.
- How severe is CVE-2025-66448?
- CVE-2025-66448 has a CVSS 3.x base score of 7.1, rated high severity. It is exploitable over network with high attack complexity, requires low privileges and user interaction. Impact on confidentiality is high, integrity high, and availability high.
- Is CVE-2025-66448 being actively exploited?
- It is not currently listed in CISA's KEV catalog. Its EPSS exploit-prediction score is 1% (44th percentile), an estimate of the probability of exploitation in the next 30 days.
- What products are affected by CVE-2025-66448?
- CVE-2025-66448 affects Vllm. See the affected-products list for the exact vulnerable versions.
- How do I fix CVE-2025-66448?
- Review the linked vendor and NVD advisories for patched versions and mitigations, then upgrade or apply the recommended workaround. Given its high severity, prioritise patching exposed systems.
- Does CVE-2025-66448 have an EU (EUVD) identifier?
- Yes. CVE-2025-66448 is tracked in the ENISA EU Vulnerability Database (EUVD) as EUVD-2025-200115.
- When was CVE-2025-66448 published?
- CVE-2025-66448 was published on 2025-12-01 and last updated on 2026-06-17.
References
- https://github.com/vllm-project/vllm/commit/ffb08379d8870a1a81ba82b72797f196838d0c86
- https://github.com/vllm-project/vllm/pull/28126
- https://github.com/vllm-project/vllm/security/advisories/GHSA-8fr4-5q9j-m8gm
Affected products (1)
- cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:*
More vulnerabilities in Vllm
- CVE-2025-32444 — Critical (CVSS 10.0): vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Versions starting from 0.6.5 and…
- CVE-2026-22778 — Critical (CVSS 9.8): vLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.1, when an invalid…
- CVE-2025-47277 — Critical (CVSS 9.8): vLLM, an inference and serving engine for large language models (LLMs), has an issue in versions 0.6.5 through 0.8.4…
- CVE-2024-11041 — Critical (CVSS 9.8): vllm-project vllm version v0.6.2 contains a vulnerability in the MessageQueue.dequeue() API function. The function uses…
- CVE-2026-48746 — Critical (CVSS 9.1): vLLM is an inference and serving engine for large language models (LLMs). From 0.3.0 until 0.22.0, a vulnerability in…
- CVE-2025-29783 — Critical (CVSS 9.0): vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. When vLLM is configured to use…
Other CWE-94 (Code Injection) vulnerabilities
- CVE-2026-57624 — Critical (CVSS 10.0): Unauthenticated Remote Code Execution (RCE) in Blocksy Companion Pro <= 2.1.46 versions.
- CVE-2026-10134 — Critical (CVSS 10.0): IBM Langflow OSS 1.0.0 through 1.9.3 allows an attacker to read every secret available to the Langflow process, read…
- CVE-2026-53576 — Critical (CVSS 10.0): Kestra is an open-source, event-driven orchestration platform. Prior to 1.0.45 and 1.3.21, the authentication filter…
- CVE-2026-10561 — Critical (CVSS 10.0): IBM Langflow OSS 1.0.0 through 1.9.3 has an vulnerability due to an improper isolation of Python execution combined…
- CVE-2026-25470 — Critical (CVSS 10.0): Improper Control of Generation of Code ('Code Injection') vulnerability in ACPT ACPT (Pro) - Custom Post Types Plugin…
- CVE-2026-48836 — Critical (CVSS 10.0): Unauthenticated Remote Code Execution (RCE) in Easy Invoice <= 2.1.19 versions.