Machine-written summary
Summary written by Claude Haiku from the source advisory. Not reviewed by a person.
The vLLM package, versions 0.7.0 up to 0.9.0 (not including 0.9.0), hashes PIL images using only raw pixel bytes, so images with different sizes or modes but identical pixel data can produce the same hash. The source does not describe a remote attack path in detail, but it says this can cause incorrect cache hits and possible data leakage. Upgrade vLLM to version 0.9.0 or later; no workaround is listed.
Not analysed yet
We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.
Severity and exploitation
| Score | Rating | CVSS | Vector | Source |
|---|---|---|---|---|
| 4.2 | Medium | 3.1 | CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:L/I:N/A:L | CNA (GitHub_M) |
- CISA KEV Not listed.
- EPSS 0.0034, the 27th percentile, on 5 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.
Timeline
- 31 Oct 2026, 14:40 UTC First seen by agentprime
- 31 Oct 2026, 14:40 UTC Confirmed by OSV
Sources
- GitHub Advisory Database: GHSA-c65p-x677-fgj6
- OSV: GHSA-c65p-x677-fgj6
- github.com/vllm-project/vllm/security/advisories/GHSA-c65p-x677-fgj6
- nvd.nist.gov/vuln/detail/CVE-2025-46722
- github.com/vllm-project/vllm/pull/17378
- github.com/vllm-project/vllm/commit/99404f53c72965b41558aceb1bc2380875f5d848
- github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2025-43.yaml
- github.com/vllm-project/vllm