CVE-2026-22778CRITICAL 9.8EPSS p60.9%

CVE-2026-22778CVE-2026-22778

Description

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.

Scoring

CVSS 3.19.8 (CRITICAL)
VectorCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
EPSS1.08% probability of exploitation · percentile 60.9% · 2026-06-19T12:03:05Z
Published2026-02-02
Last modified2026-02-23

Underlying weaknesses· 1

CWE-532

References

  1. https://github.com/vllm-project/vllm/pull/31987
  2. https://github.com/vllm-project/vllm/pull/32319
  3. https://github.com/vllm-project/vllm/releases/tag/v0.14.1
  4. https://github.com/vllm-project/vllm/security/advisories/GHSA-4r2x-xpjr-7cvv

1

TypeTargetConfidenceTier
WeaknessInsertion of Sensitive Information into Log Filecwe-5320%live

Related by meaning· 6

Nearest entities by semantic similarity across the cs-graph corpus.

CVE
CVE-2025-62164
CVE
CVE-2026-22807
CVE
CVE-2026-27893
CVE
CVE-2026-25960
CVE
CVE-2025-24357
CVE
CVE-2025-29783
Sourced from NVD + FIRST.org EPSS. Curated for EU compliance use cases by Adam Lundqvist, Founder at SQUR.