INCIDENT MAP

Browse / AI Framework and Platform Flaws / CVE-2025-46722

vLLM has a Weakness in MultiModalHasher Image Hashing Implementation

CVE-2025-46722 · 2025-05-28 moderate

Class
AI Framework and Platform Flaws (filed by a model that read the advisory)
Technique
Other weaknesses
Also involves
Sensitive Data Disclosure
Affected
vllm (PyPI)
CVSS
CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:L/I:N/A:L
EPSS
0.3% chance of exploitation in 30 days

How it works, in short

The image hash uses only raw pixel bytes without dimensions or mode, so different images can share a hash and receive each other's cached results. (Our summary, written by a model. The source text is below.)

Details

Summary

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.

Details

Recommendation

In the serialize_item method, serialization of Image.Image objects should include not only pixel data, but also all critical metadata—such as dimensions (size), color mode (mode), format, and especially the info dictionary. The info dictionary is particularly important in palette-based images (e.g., mode 'P'), where the palette itself is stored in info. Ignoring info can result in hash collisions between visually distinct images with the same pixel bytes but different palettes or metadata. This can lead to incorrect cache hits or even data leakage.

Summary:
Serializing only the raw pixel data is insecure. Always include all image metadata (size, mode, format, info) in the hash calculation to prevent collisions, especially in cases like palette-based images.

Impact for other modalities
For the influence of other modalities, since the video modality is transformed into a multi-dimensional array containing the length, width, time, etc. of the video, the same problem exists due to the incorrect sequence of numpy as well.

For audio, since the momo function is not enabled in librosa.load, the loaded audio is automatically encoded into single channels by librosa and returns a one-dimensional array of numpy, thus keeping the structure of numpy fixed and not affected by this issue.

Fixes

References

Source: GitHub Advisory Database, used under CC BY 4.0.