Memory Leak Affecting vllm-cu130 package, versions <0.30.0-r0


Severity

Recommended
0.0
high
0
10

Snyk's Security Team recommends NVD's CVSS assessment. Learn more

Threat Intelligence

EPSS
0.63% (49th percentile)

Do your applications use this vulnerable package?

In a few clicks we can analyze your entire application and see what components are vulnerable in your application, and suggest you quick fixes.

Test your applications
  • Snyk IDSNYK-MINIMOSLATEST-VLLMCU130-20267699
  • published30 Sept 2026
  • disclosed21 Sept 2026

Introduced: 21 Sep 2026

NewCVE-2026-94627  (opens in a new tab)
CWE-401  (opens in a new tab)

How to fix?

Upgrade Minimos:latest vllm-cu130 to version 0.30.0-r0 or higher.

NVD Description

Note: Versions mentioned in the description apply only to the upstream vllm-cu130 package and not the vllm-cu130 package as distributed by Minimos. See How to fix? for Minimos:latest relevant fixed versions and status.

vLLM Mooncake connector through 0.29.0 fails to properly manage GPU KV cache block ownership when concurrent child requests share a single transfer ID in prefill/decode disaggregated deployments. Attackers can trigger GPU memory exhaustion by submitting completion requests with multiple prompts, causing orphaned KV cache blocks to accumulate until process restart and eventually preventing legitimate requests from executing.

CVSS Base Scores

version 3.1