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.
Weakness
The product deserializes untrusted data without sufficiently ensuring that the resulting data will be valid.
Affected Software
| Name | Vendor | Start Version | End Version |
|---|---|---|---|
| Vllm | Vllm | * | 0.32.0 (excluding) |
Potential Mitigations
- Make fields transient to protect them from deserialization.
- An attempt to serialize and then deserialize a class containing transient fields will result in NULLs where the transient data should be. This is an excellent way to prevent time, environment-based, or sensitive variables from being carried over and used improperly.