VulnerabilityAnalyzed

CVE-2025-46722

This may lead to hash collisions, incorrect cache hits, and even data leakage or security risks.

HIGH 7.3EPSS 0.32%

Does this matter?

High impact if exploited, but EPSS currently rates exploitation as unlikely (0.32%). Schedule it in the normal patch cycle and watch for a rise in EPSS or a public exploit.

Description

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.

CVSS 3.1: 7.3 HIGHCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L
EPSS: 0.32% probability · 24th percentile
CISA KEV: Not listed
Weakness: CWE-1023, CWE-1288
Affected: vllm/vllm
Source: security-advisories@github.com

References

https://github.com/vllm-project/vllm/commit/99404f53c72965b41558aceb1bc2380875f5d848Patch
https://github.com/vllm-project/vllm/pull/17378Issue Tracking, Patch
https://github.com/vllm-project/vllm/security/advisories/GHSA-c65p-x677-fgj6Vendor Advisory

Source: NVD record, EPSS from FIRST.org, KEV from CISA. Refreshed daily.