CVE-2026-22773
Medium
Description
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.
SeverityMedium
CVSS6.5
Exploit probability—
ProductVllm
PublishedJan 10, 2026
First seenJan 10, 2026
Weakness typeCWE-770Allocation of resources without limits
How to fix
Update
- vllm vllm0.12.0 or later
- vllm0.12.0 or later
Affected products
| Product | Affected versions | Fixed in |
|---|---|---|
| vllm vllm | 0.6.4 ≤ v < 0.12.0 | 0.12.0 |
| vllmPyPI | 0.6.4 ≤ v < 0.12.0 | 0.12.0 |
Check your version
Choose the product and enter the version you run to see whether it falls in the affected range.
In the Intelligence 360 plan
Attack Surface Management
See the assets tied to your domains, the technologies they run and known vulnerabilities in one list.