Vllm-Project
AI Threat Alert tracks 9 known AI/ML vulnerabilities affecting Vllm-Project products — each enriched with CVSS severity, EPSS exploit probability, patch status, and CISO-grade analysis. Browse every Vllm-Project CVE below, sorted by severity and recency.
| Severity | CVE | Headline | Package | CVSS |
|---|---|---|---|---|
| CRITICAL | CVE-2024-9053 | vllm: RCE via unsafe pickle deserialization in RPC server | vllm | 9.8 |
| MEDIUM | CVE-2026-78684 | vLLM: unauthenticated DoS via DeepStream decode bypass | vllm | 5.3 |
| MEDIUM | CVE-2026-69147 | vLLM: request-selected video backend exhausts GPU memory | vllm | 6.5 |
| MEDIUM | CVE-2026-105775 | vLLM: OOB read in Mamba2 mixer enables remote DoS | vLLM | 4.3 |
| HIGH | CVE-2026-115915 | vLLM: image decoder abuse enables RCE/DoS | vllm | 8.1 |
| MEDIUM | CVE-2026-115916 | vLLM: off-by-one token ID check crashes engine (DoS) | vllm | 6.5 |
| MEDIUM | CVE-2026-115914 | vLLM: zero-sum OCR image crashes engine (DoS) | vllm | 6.5 |
| LOW | CVE-2026-115917 | vLLM: wildcard CORS lets web pages reach local servers | vllm | 3.4 |
| MEDIUM | CVE-2026-115918 | vLLM: chat template bypass enables DoS via /v1/responses | vllm | 6.5 |
Frequently asked questions
How many known vulnerabilities affect Vllm-Project?
9 AI/ML CVEs affecting Vllm-Project products are tracked, sourced from NVD and GitHub Advisory.
What Vllm-Project products are affected?
The CVEs below map to the Vllm-Project AI/ML packages and tools tracked by AI Threat Alert; open any CVE to see the affected components and versions.
Where does the Vllm-Project vulnerability data come from?
Data is sourced from NVD and GitHub Advisory, then enriched with CVSS severity, EPSS exploit probability, and patch status for each CVE.
How can I monitor Vllm-Project for new vulnerabilities?
AI Threat Alert tracks Vllm-Project continuously; a Pro subscription adds breaking alerts when new CVEs affecting Vllm-Project are published.
How do I assess Vllm-Project's security exposure?
Each CVE below carries CVSS severity and exploitation signals, so you can review the highest-severity Vllm-Project issues first and judge the exposure for your stack.