CVE-2025-46722
CVE-2025-46722 is a medium-severity vulnerability in Vllm with a CVSS 3.x base score of 4.2. 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-1023.
Key facts
- Severity: Medium (CVSS 3.x base score 4.2)
- EPSS exploit prediction: 0% (18th percentile)
- Actively exploited: Not listed in CISA KEV
- EU (EUVD) id: EUVD-2025-16188
- Weakness: CWE-1023
- Affected product: Vllm
- Published:
- Last modified:
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.
Frequently asked questions
- What is CVE-2025-46722?
- 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.
- How severe is CVE-2025-46722?
- CVE-2025-46722 has a CVSS 3.x base score of 4.2, rated medium severity. It is exploitable over network with high attack complexity, requires low privileges and no user interaction. Impact on confidentiality is low, integrity none, and availability low.
- Is CVE-2025-46722 being actively exploited?
- It is not currently listed in CISA's KEV catalog. Its EPSS exploit-prediction score is 0% (18th percentile), an estimate of the probability of exploitation in the next 30 days.
- What products are affected by CVE-2025-46722?
- CVE-2025-46722 affects Vllm. See the affected-products list for the exact vulnerable versions.
- How do I fix CVE-2025-46722?
- Review the linked vendor and NVD advisories for patched versions and mitigations, then upgrade or apply the recommended workaround.
- Does CVE-2025-46722 have an EU (EUVD) identifier?
- Yes. CVE-2025-46722 is tracked in the ENISA EU Vulnerability Database (EUVD) as EUVD-2025-16188.
- When was CVE-2025-46722 published?
- CVE-2025-46722 was published on 2025-05-29 and last updated on 2026-06-17.
References
- https://github.com/vllm-project/vllm/commit/99404f53c72965b41558aceb1bc2380875f5d848
- https://github.com/vllm-project/vllm/pull/17378
- https://github.com/vllm-project/vllm/security/advisories/GHSA-c65p-x677-fgj6
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-1023 vulnerabilities
- CVE-2026-4599 — Critical (CVSS 9.1): Versions of the package jsrsasign from 7.0.0 and before 11.1.1 are vulnerable to Incomplete Comparison with Missing…
- CVE-2025-62000 — High (CVSS 7.1): BullWall Ransomware Containment may not always detect an encrypted file. This issue affects a specific file inspection…
- CVE-2026-53839 — Medium (CVSS 6.5): OpenClaw before 2026.5.7 contains a hostname validation vulnerability in retry endpoint checks that allows matching…
- CVE-2025-55333 — Medium (CVSS 6.1): Incomplete comparison with missing factors in Windows BitLocker allows an unauthorized attacker to bypass a security…
- CVE-2026-7473 — Medium (CVSS 5.8): On affected platforms running Arista EOS where a tunnel decapsulation configuration—such as VXLAN (Virtual Extensible…
- CVE-2026-48587 — Low (CVSS 3.1): An issue was discovered in Django 5.2 before 5.2.15 and 6.0 before 6.0.6. `django.utils.cache.has_vary_header()` in…