CVE-2025-62372Medium· 6.5▾ SunlitvLLM vulnerable to DoS with incorrect shape of multimodal embedding inputs
▾ Sunlit zone — Low / medium · no exploitation signal
impact 35.8 · likelihood 0.1 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
Exploit-prediction probability, daily snapshots since Jul 8.
Disclosure to exploitation, from the record and what we observed since indexing it.
Disclosed via OSV
0.3%
0.3% → 0.4%
Last analysed / modified upstream
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)
Using image embeddings as an example:
inputs_embeds (mismatched shape)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).
--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.vllm >= 0.5.5, < 0.11.1Upgrade to a patched release:
vllm 0.11.1Connected by shared product, vendor, weakness, or advisory.
CVE-2025-62164High· 8.8vLLM deserialization vulnerability leading to DoS and potential RCE
CVE-2026-73558Medium· 5.3vLLM is an inference and serving engine for large language models
CVE-2026-73560Medium· 6.5vLLM is an inference and serving engine for large language models
CVE-2026-71486Medium· 4.3vLLM is an inference and serving engine for large language models
CVE-2026-73555Medium· 5.3vLLM is an inference and serving engine for large language models
CVE-2026-73556Medium· 5.3vLLM is an inference and serving engine for large language models