vLLM: Phi-4 multimodal audio encoder materializes and retains full attention matrices: a single max-length audio OOM-crashes the vLLM engine

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

• CISA KEV Not listed.
• EPSS 0.0033, the 26th percentile, on 5 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

Timeline

1. 26 Oct 2026, 16:56 UTC Confirmed by the CVE List
2. 26 Oct 2026, 17:17 UTC Confirmed by NVD
3. 31 Oct 2026, 14:07 UTC First seen by agentprime
4. 31 Oct 2026, 14:07 UTC Confirmed by OSV

Sources

• GitHub Advisory Database: GHSA-8397-8qf6-f2gm
• OSV: GHSA-8397-8qf6-f2gm
• github.com/vllm-project/vllm/security/advisories/GHSA-8397-8qf6-f2gm
• nvd.nist.gov/vuln/detail/CVE-2026-115913
• github.com/vllm-project/vllm/pull/60614
• github.com/vllm-project/vllm/commit/3d52578e13454de0bb8352cac988347cdcc89928
• github.com/vllm-project/vllm
• github.com/vllm-project/vllm/releases/tag/v0.32.0

Facts on this page come from the GitHub Advisory Database (CC-BY 4.0), OSV and the other sources listed, as published. We don’t reproduce full advisory descriptions, which can contain exploit code.