CVE-2026-22773
vLLM is vulnerable to DoS in Idefics3 vision models via image payload with ambiguous dimensions
6.5
MEDIUM
CVSS 3.1
EPSS 0.02%
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.
How to fix CVE-2026-22773
To remediate CVE-2026-22773, upgrade the affected package to a fixed version below.
- —upgrade to 0.12.0 or later
- —upgrade to 0.12.0 or later
Is CVE-2026-22773 being exploited?
Low — EPSS is 0.0%, meaning exploitation activity has not been observed at scale.
Affected packages (2)
- >= 0.6.4, < 0.12.0
- >= 0.6.4, < 0.12.0
CVSS scores
| Source | Version | Severity | Vector |
|---|---|---|---|
| osv | CVSS 3.1 | MEDIUM6.5 | CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H |