CVEvulnerability
CVE-2026-115912
8.8 HIGH
Official page ↗
- Status
- Analyzed
- Published
- 2026-10-26
- Last modified
- 2026-10-28
- Assigner
- [email protected]
- CVSS 3.1 ([email protected])
- 7.5 HIGH — CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H
- CVSS 3.1 ([email protected])
- 8.8 HIGH — CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
- Weaknesses
- CWE-502
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 execution even when insecure serialization is disabled. This issue is fixed in version 0.32.0.
Affected configurations
• vllm vllm (< 0.32.0)
Related
Weakness · CWE (1)
- CWE-502 Deserialization of Untrusted Data weakness
References
- https://github.com/vllm-project/vllm/commit/3d5625cc3c2474994107047d7a6ecb8e4008f9b0 Patch
- https://github.com/vllm-project/vllm/pull/59357 Patch
- https://github.com/vllm-project/vllm/releases/tag/v0.32.0 Release Notes
- https://github.com/vllm-project/vllm/security/advisories/GHSA-6757-hggm-hp2w Patch, Vendor Advisory