CVE-2026-22778
EUVD-2026-527702.02.2026, 23:16
vLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.1, when an invalid image is sent to vLLM's multimodal endpoint, PIL throws an error. vLLM returns this error to the client, leaking a heap address. With this leak, we reduce ASLR from 4 billion guesses to ~8 guesses. This vulnerability can be chained a heap overflow with JPEG2000 decoder in OpenCV/FFmpeg to achieve remote code execution. This vulnerability is fixed in 0.14.1.Enginsight
Affected Products (NVD)
| Vendor | Product | Version |
|---|---|---|
| vllm | vllm | 0.8.3 ≤ 𝑥 < 0.14.1 |
𝑥
= Vulnerable software versions
Early Detection
Affected products identified ahead of NVD analysis through intelligence sources.
| Vendor | Product | Version | Source |
|---|---|---|---|
| Red Hat | Red Hat AI Inference Server 3.2 | 1772160593 ≤ 𝑥 < * | ADP |
| Red Hat | Red Hat AI Inference Server 3.2 | 1772160625 ≤ 𝑥 < * | ADP |
| Red Hat | Red Hat AI Inference Server 3.3 | 1782352919 ≤ 𝑥 < * | ADP |
| Red Hat | Red Hat AI Inference Server 3.3 | 1782353093 ≤ 𝑥 < * | ADP |
| Red Hat | Red Hat AI Inference Server 3.3 | 1782352847 ≤ 𝑥 < * | ADP |
| Red Hat | Red Hat OpenShift AI 2.25 | 1772093436 ≤ 𝑥 < * | ADP |
| Red Hat | Red Hat OpenShift AI 2.25 | 1772093276 ≤ 𝑥 < * | ADP |
| Red Hat | Red Hat OpenShift AI 2.25 | 1772093278 ≤ 𝑥 < * | ADP |
| Red Hat | Red Hat OpenShift AI 2.25 | 1772093237 ≤ 𝑥 < * | ADP |
| Red Hat | Red Hat OpenShift AI 3.3 | 1778264363 ≤ 𝑥 < * | ADP |
| Red Hat | Red Hat OpenShift AI 3.3 | 1770956034 ≤ 𝑥 < * | ADP |
Common Weakness Enumeration
- CWE-532 - Insertion of Sensitive Information into Log FileInformation written to log files can be of a sensitive nature and give valuable guidance to an attacker or expose sensitive user information.
- CWE-209 - Generation of Error Message Containing Sensitive InformationThe software generates an error message that includes sensitive information about its environment, users, or associated data.
References