CVE-2026-53923Network attack vector

Cross-tenant tensor disclosure via truncated CUDA counts

Published Jun 22, 2026 · Updated Jun 22, 2026

Information disclosure in vLLM 0.5.5 and later, before 0.23.1rc0, allows remote attackers to expose residual GPU tensor data via crafted GGUF models. The GGUF dequantization path passes 64-bit tensor element counts into 32-bit CUDA kernel parameters, so truncated launch bounds leave part of each torch::empty output tensor unwritten and populated with stale GPU memory. Exploitation requires a victim to load an attacker-crafted GGUF model with more than INT_MAX tensor elements in a multi-tenant inference deployment, where residual data from other users' requests can enter model computation.

CVSS severity5.3
Medium
EPSS probability0.48%
Next 30 days · Sep 16, 2026
Known exploitationUnconfirmed
Based on sourced intelligence
Hinoki checkNot available
Coverage for this vulnerability

See if you're affected

Explore vulnerability checks for your environment with Hinoki.

Book a demo

Summary

What happened

Information disclosure in vLLM 0.5.5 and later, before 0.23.1rc0, allows remote attackers to expose residual GPU tensor data via crafted GGUF models. The GGUF dequantization path passes 64-bit tensor element counts into 32-bit CUDA kernel parameters, so truncated launch bounds leave part of each torch::empty output tensor unwritten and populated with stale GPU memory. Exploitation requires a victim to load an attacker-crafted GGUF model with more than INT_MAX tensor elements in a multi-tenant inference deployment, where residual data from other users' requests can enter model computation.

The record

CVE
CVE-2026-53923
Published
Jun 22, 2026
Updated
Jun 22, 2026
Vendor
vLLM Project
Product
vLLM
Classifications
CWE-200, CWE-681, T1204.002
Attack vector
network
Privileges
unauthenticated

Timeline

How it unfolded

  1. Jun 22, 2026CVE publishedPublication date reported by the CVE source.
  2. Jun 22, 2026Record updatedLatest update available in the CVE record.

Exploitability

Present is not the same as exploitable

Compare your product and version with the public record. A matching version still requires validation against your environment.

Is a vulnerable build present?

Compare these published version ranges with your installed build and any vendor patches.

  1. Affected versionversion=>= 0.5.5, < 0.23.1rc0

What conditions does exploitation require?

Attack vectornetwork
Required privilegesunauthenticated

What is affected?

vLLM Project · vLLMversion=>= 0.5.5, < 0.23.1rc0

Published CVSS scores

7.5NIST NVDCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N
5.3GitHubCVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:P/VC:L/VI:L/VA:N/SC:N/SI:N/SA:N

CVSS describes severity. EPSS estimates exploitation probability.

Attacks

What attackers are doing with it

Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.

Daily unique IPsNo honeypot observations are available for this CVE in the selected window.

No observations available

Sep 10, 2026Sep 16, 2026
Latest reporting daySep 16, 2026
Latest daily unique IPsUnavailable
Prior 30-day averageUnavailable
SourceShadowserver honeypots (KEV)
Vectornetwork
Privilegesunauthenticated
Known exploitationUnconfirmed
Public exploitUnconfirmed

Weakness, pattern, technique

CWE-200Exposure of Sensitive Information to an Unauthorized Actor
CWE-681Incorrect Conversion between Numeric Types
T1204.002Malicious File

Public exploit references

No public exploit references are available in this record.

Labels summarize the accepted research assessment. They do not indicate a test against your environment.

Technologies

Your stack

See the directory against your own environment.

Your stack

Check the software in your environment

Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.

Book a demo