Machine-written summary
Summary written by Claude Haiku from the source advisory. Not reviewed by a person.
vLLM versions from 0.7.3 before 0.31.0 let API clients pass arbitrary per-request mm_processor_kwargs, which are forwarded to multimodal image and video processors, so a remote client can control values such as image size, patch count and frame count. A single crafted request can exhaust memory and terminate the API server or worker process, affecting all users of that instance; no privileges are required when the optional API key is not set. Upgrade to vLLM 0.31.0 or later. Until then, operators can limit access to trusted clients behind an external authentication layer.
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 |
|---|---|---|---|---|
| 8.7 | High | 4.0 | CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N | CNA (VulnCheck) |
- CISA KEV Not listed.
- EPSS 0.0042, the 36th percentile, on 5 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.
Timeline
- 21 Oct 2026, 01:17 UTC Confirmed by NVD
- 31 Oct 2026, 14:08 UTC First seen by agentprime
- 31 Oct 2026, 14:08 UTC Confirmed by OSV
Sources
- GitHub Advisory Database: GHSA-gx7p-2j49-hfq4
- OSV: GHSA-gx7p-2j49-hfq4
- github.com/vllm-project/vllm/security/advisories/GHSA-gx7p-2j49-hfq4
- nvd.nist.gov/vuln/detail/CVE-2026-114055
- github.com/vllm-project/vllm/pull/58830
- github.com/vllm-project/vllm/commit/31f2e70cd3e403aaa471fb7fe3d33a7207aaf3cc
- github.com/vllm-project/vllm
- github.com/vllm-project/vllm/releases/tag/v0.31.0
- www.vulncheck.com/advisories/vllm-0.7.3-before-0.31.0-dos-via-unvalidated-mm-processor-kwargs