Advisory details:
pypa / vllm/PYSEC-2026-4345
| Advisory ID | pypa/vllm/PYSEC-2026-4345 |
| Aliases |
CVE-2026-115912
GHSA-6757-hggm-hp2w PYSEC-2026-4345 |
| Summary | 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. |
| Status | Published |
| Exploitability | 0.5 |
| Weighted Severity | 7.9 |
| Risk | 4.0 |
| Affected and Fixed Packages | Package Details |
| Introduced and Fixed Package Commit Patches | Package Commit Patches Details |
| There are no known CWE. |
| System | Score | Found at |
|---|---|---|
| cvssv3.1 | 8.8 | https://github.com/pypa/advisory-database/blob/main/vulns/vllm/PYSEC-2026-4345.yaml |
No EPSS data available for this advisory.
| Patch URL | |
|---|---|
| There are no known patches. | |
| Attack Vector (AV) | Attack Complexity (AC) | Privileges Required (PR) | User Interaction (UI) | Scope (S) | Confidentiality Impact (C) | Integrity Impact (I) | Availability Impact (A) |
|---|---|---|---|---|---|---|---|
network adjacent_network local physical |
low high |
none low high |
none required |
unchanged changed |
high low none |
high low none |
high low none |
There are no SSVC decisions available.