Machine-written summary
Summary written by Claude Haiku from the source advisory. Not reviewed by a person.
Affected versions of the vllm package before 0.9.0 are vulnerable to a timing side-channel in prefix caching: when a prompt's prefix chunks match cached content, time to first token drops measurably. An attacker who shares the same backend and can send requests and measure response timing could infer whether their guesses match a victim's cached prompt, potentially leaking private information. Upgrade to vllm 0.9.0 or later.
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 |
|---|---|---|---|---|
| 2.6 | Low | 3.1 | CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:U/C:L/I:N/A:N | CNA (GitHub_M) |
- CISA KEV Not listed.
- EPSS 0.0043, the 37th percentile, on 6 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-4qjh-9fv9-r85r
- OSV: GHSA-4qjh-9fv9-r85r
- github.com/vllm-project/vllm/security/advisories/GHSA-4qjh-9fv9-r85r
- nvd.nist.gov/vuln/detail/CVE-2025-46570
- github.com/vllm-project/vllm/pull/17045
- github.com/vllm-project/vllm/commit/77073c77bc2006eb80ea6d5128f076f5e6c6f54f
- github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2025-53.yaml
- github.com/vllm-project/vllm