LLMMedium
CVE-2026-22773: vLLM is an inference and serving engine for large language models (LLMs). In versions from 0.6.4 to …
- Occurred
- Jan 10, 2026
- Published
- Jan 10, 2026
- Source
- NVD CVE
Summary
vLLM is an inference and serving engine for large language models (LLMs). In versions from 0.6.4 to before 0.12.0, users can crash the vLLM engine serving multimodal models that use the Idefics3 vision model implementation by sending a specially crafted 1x1 pixel image. This causes a tensor dimension mismatch that results in an unhandled runtime error, leading to complete server termination. This issue has been patched in version 0.12.0.
Tags
- CVE
- CVE-2026-22773
- cvss:MEDIUM
Classification
Category confidence: 95%
Severity confidence: 90%
Rationale: CVE describes a denial-of-service vulnerability in an LLM serving engine causing server crashes.
Classifier: Qwen/Qwen3.5-35B-A3B-FP8