vLLM vulnerable to remote code execution via transformers_utils/get_config

Summary

vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.11.1, vllm has a critical remote code execution vector in a config class named Nemotron_Nano_VL_Config. When vllm loads a model config that contains an auto_map entry, the config class resolves that mapping with get_class_from_dynamic_module(...) and immediately instantiates the returned class. This fetches and executes Python from the remote repository referenced in the auto_map string. Crucially, this happens even when the caller explicitly sets trust_remote_code=False in vllm.transformers_utils.config.get_config. In practice, an attacker can publish a benign-looking frontend repo whose config.json points via auto_map to a separate malicious backend repo; loading the frontend will silently run the backend’s code on the victim host. This vulnerability is fixed in 0.11.1.

Timeline

1. Dec 1, 2025, 10:45 PM UTCCVE published
2. Dec 2, 2025, 2:14 PM UTCCISA SSVC assessment

   Exploitation: none · Automatable: yes · Technical impact: total

3. Dec 2, 2025, 2:14 PM UTCRecord last updated

Affected versions

• vllm-project vllm

  fixed in 0.11.1

• vllmPyPI

  fixed in 0.11.1

References 3