CVE-2025-32444
Description
vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Versions starting from 0.6.5 and prior to 0.8.5, having vLLM integration with mooncake, are vulnerable to remote code execution due to using pickle based serialization over unsecured ZeroMQ sockets. The vulnerable sockets were set to listen on all network interfaces, increasing the likelihood that an attacker is able to reach the vulnerable ZeroMQ sockets to carry out an attack. vLLM instances that do not make use of the mooncake integration are not vulnerable. This issue has been patched in version 0.8.5.
How to fix
- vllm vllm0.8.5 or later
- vllm0.8.5 or later
Affected products
| Product | Affected versions | Fixed in |
|---|---|---|
| vllm vllm | 0.6.5 ≤ v < 0.8.5 | 0.8.5 |
| vllmPyPI | 0.6.5 ≤ v < 0.8.5 | 0.8.5 |
Check your version
Choose the product and enter the version you run to see whether it falls in the affected range.
Attack Surface Management
See the assets tied to your domains, the technologies they run and known vulnerabilities in one list.