AIIC AI Intelligence Centre

SOURCE-LINKED INTELLIGENCE

GHSA-pmqf-x6x8-p7qw: vLLM vulnerable to DoS with incorrect shape of multimodal embedding inputs

OSV AI package advisories · observation · Nov 20, 2025 · UTC

### Summary Users can crash the vLLM engine serving multimodal models by passing multimodal embedding inputs with correct `ndim` but incorrect `shape` (e.g. hidden dimension is wrong), regardless of whether the model is intended to support such inputs (as defined in the Supported Models page). The issue has existed ever since we added support for image embedding inputs, i.e. #6613 (released in v0.5.5) ### Details Using image embeddings as an example: - For models that support image embedding inputs, the engine crashes when scattering the embeddings to `inputs_embeds` (mismatched shape) - For m

Read original source ↗ Open in workspace

recordType
vulnerability
status
active
evidenceStatus
reported
region
Global

Evidence & attribution

First collected: 2026-09-20T22:31:48.298Z. This is not the publication date.