UpHeron
All vulnerabilities

Servers and applications

CVE-2026-115912

Public multimodal SHM handle forgery reaches worker-side `pickle.loads()` with insecure serialization disabled

Severity
High 8.8
Published
20 Oct 2026
Last changed
28 Oct 2026
Added to UpHeron
3 Nov 2026
Fixed in
0.32.0
Also known as
GHSA-6757-hggm-hp2w
Sources
NVD, CVE Program, Project advisories (GitHub)

Description

vLLM is an inference and serving engine for large language models. Prior to 0.32.0, an authenticated caller can place forged shared-memory address and monotonic_id values in public multimodal features.kwargs_data. In scale-out deployments with shared-memory multimodal caching enabled, workers trust those values and pass attacker-selected bytes to pickle.loads(), allowing arbitrary Python…

Affected software

SoftwareAffected versions
vLLMServers and applications · cna:vllm-project|vllm
  • Every version before 0.32.0
vLLMServers and applications · cpe:a:vllm:vllm
  • Every version before 0.32.0
vLLMServers and applications · github:vllm-project/vllm
  • Every version before 0.32.0

References

  • NVD https://nvd.nist.gov/vuln/detail/CVE-2026-115912
  • CVE.org https://www.cve.org/CVERecord?id=CVE-2026-115912
  • GitHub Advisory Database https://github.com/advisories/GHSA-6757-hggm-hp2w
  • github.com https://github.com/vllm-project/vllm/security/advisories/GHSA-6757-hggm-hp2w