CVE-2025-62372
vLLM vulnerable to DoS with incorrect shape of multimodal embedding inputs
Description
### 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 models that don't support image embedding inputs, the engine crashes when validating the inputs inside `get_input_embeddings` (validation fails). This happens because we only validate `ndim` of the tensor, but not the full shape, in input processor (via `MultiModalDataParser`). ### Impact - Denial of service by crashing the engine ### Mitigation - Use API key to limit access to trusted users. - Set `--limit-mm-per-prompt` to 0 for all non-text modalities to ban multimodal inputs, which includes multimodal embedding inputs. However, the model would then only accept text, defeating the purpose of using a multi-modal model. ### Resolution - https://github.com/vllm-project/vllm/pull/27204
How to fix CVE-2025-62372
To remediate CVE-2025-62372, upgrade the affected package to a fixed version below.
- —upgrade to 0.11.1 or later
Is CVE-2025-62372 being exploited?
Low — EPSS is 0.1%, meaning exploitation activity has not been observed at scale.
Affected packages (1)
- >= 0.5.5, < 0.11.1