CVE-2026-115912
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.
Does this matter?
High impact if exploited, but EPSS currently rates exploitation as unscored. Schedule it in the normal patch cycle and watch for a rise in EPSS or a public exploit.
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.
- CVSS 3.1
- 7.5 HIGHCVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H
- EPSS
- No score yet
- Public exploits
- None in Exploit-DB
- CISA KEV
- Not listed
- Weakness
- CWE-502
- Source
- security-advisories@github.com
References
Source: NVD record, EPSS from FIRST.org, KEV from CISA, exploits from Exploit-DB. Refreshed daily. Download this record as JSON.