CVE-2025-46722

描述

vLLM is an inference and serving engine for large language models (LLMs). In versions starting from 0.7.0 to before 0.9.0, in the file vllm/multimodal/hasher.py, the MultiModalHasher class has a security and data integrity issue in its image hashing method. Currently, it serializes PIL.Image.Image objects using only obj.tobytes(), which returns only the raw pixel data, without including metadata such as the image’s shape (width, height, mode). As a result, two images of different sizes (e.g., 30x100 and 100x30) with the same pixel byte sequence could generate the same hash value. This may lead to hash collisions, incorrect cache hits, and even data leakage or security risks. This issue has been patched in version 0.9.0.

如何修補 CVE-2025-46722

要修補 CVE-2025-46722,請將受影響套件升級到下列已修補版本。

• —升級至 0.9.0 或更新版本
• —升級至 99404f53c72965b41558aceb1bc2380875f5d848 或更新版本

CVE-2025-46722 正在被利用嗎?

低 — EPSS 為 0.3%,目前沒有觀察到大規模利用活動。

受影響套件(2)

• >= 0.7.0, < 0.9.0
• from 0, < 99404f53c72965b41558aceb1bc2380875f5d848 | from 0, < 0.9.0

CVSS 分數

參考連結(6)

• ADVISORYnvd.nist.gov/vuln/detail/CVE-2025-46722
• PATCHgithub.com/vllm-project/vllm
• WEBgithub.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2025-43.yaml
• WEBgithub.com/vllm-project/vllm/commit/99404f53c72965b41558aceb1bc2380875f5d848