CVE Watchtower

CVE-2025-62164Improper Input Validation

NVDCVSS scored by security-advisories@github.com

Vulnerability Summary

vLLM is an inference and serving engine for large language models (LLMs). From versions 0.10.2 to before 0.11.1, a memory corruption vulnerability could lead to a crash (denial-of-service) and potentially remote code execution (RCE), exists in the Completions API endpoint. When processing user-supplied prompt embeddings, the endpoint loads serialized tensors using torch.load() without sufficient validation. Due to a change introduced in PyTorch 2.8.0, sparse tensor integrity checks are disabled by default. As a result, maliciously crafted tensors can bypass internal bounds checks and trigger an out-of-bounds memory write during the call to to_dense(). This memory corruption can crash vLLM and potentially lead to code execution on the server hosting vLLM. This issue has been patched in version 0.11.1.
Severity Level
HIGH(8.8)
Published Date
Nov 21, 2025
Last Modified
Nov 21, 2025
Exploitation Status
No confirmed exploitation yet
Root Weakness (CWE)
The product receives input or data, but it does not validate or incorrectly validates that the input has the properties that are required.
See the MITRE CWE entry for the full definition of this weakness.
The product deserializes untrusted data without sufficiently ensuring that the resulting data will be valid.
The product writes data past the end, or before the beginning, of the intended buffer.
CVSS v3.1 Base Metrics — Score 8.8 (HIGH)
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Attack VectorNetwork
Attack ComplexityLow
Privileges RequiredLow
User InteractionNone
ScopeUnchanged
ConfidentialityHigh
IntegrityHigh
AvailabilityHigh

Affected & Patched Versions

Affected Versions
Not provided by NVD for this CVE.
Patched Versions
No patched version published yet
Until a fix ships, reduce exposure:
  • Restrict access to the affected service or interface to trusted networks.
  • Disable the affected feature if you do not need it.
  • Watch the vendor advisory for a firmware or software update.

What to do

Plan remediation soon. No exploitation is confirmed yet, but the severity is high.

  1. Check whether you run the affected product and which version.
  2. No patched version is listed yet. Limit access to the affected service to trusted networks, and disable the feature if you can.
  3. Subscribe to the vendor advisory so you learn when a fix ships.

Timeline

  • Nov 21, 2025Published
  • Nov 21, 2025Record last modified

Score, severity and exploitation changes are tracked from the day Watchtower started logging them.

Get alerts for CVEs like this oneYou are reading CVE-2025-62164. Get an email when KEV listings, exploits or new CVEs match your stack.
Set up CVE alerts

Data last checked: Nov 22, 2026 18:26 UTC