Skip to content
Connect AISign inSign up free
CVE-2026-1057546.5MediumCWE-617

CVE-2026-105754: vLLM: Scale-out disaggregated multimodal transport trusts caller-supplied features

vllm-project · vllm

Affected: vllm-project · Exploitation: unlikely (under 1%)

Not in CyberSIXT coverage yet: no article names this CVE. This page shows the public record.

0.3%probability of exploitation in the next 30 days
19th percentile
050100

Higher than 19% of scored CVEs

6.5MediumCVSS 3.1
LowMedium

DescriptionCVE.org

vLLM is an inference and serving engine for large language models. Prior to 0.30.0, the /inference/v1/generate endpoint in the disaggregated scale-out path accepts caller-supplied tensors in the features.kwargs_data field, cache identifiers in the features.mm_hashes field, ranges in the features.mm_placeholders field, and wire-selected multimodal field processors without rebinding them to the active model renderer contract. Forged grid geometry, field types, or non-positive placeholder lengths can terminate the shared EngineCore; when an attacker knows or can induce a victim's content hash, forged cache hashes can poison or retrieve cross-request encoder-cache state; and dropped sparse placeholder masks can alter replayed transport semantics. This issue is fixed in version 0.30.0.

Affected products1

From the CVE record. A product or version that is not listed is unknown, not unaffected.

  • vllm-project · vllm< 0.30.0

On the Atlas

Every published CVE is on the CyberSIXT Atlas, grouped by weakness, vendor and month.

Open the Atlas
Exploitation
Unlikely (under 1%). EPSS under 1%, or no score yet.
CISA KEV
Not listed
Public check
No public nuclei template
Published
October 2026

CoverageNo articles yet

No CyberSIXT article names CVE-2026-105754 yet, so there is no coverage timeline, action brief or change log for it. Everything on this page comes from the public CVE record, CISA KEV, EPSS and the nuclei template index. Search the archive or see the CVEs in the news.

References