CVE-2025-62164

Описание

vLLM is an inference and serving engine for large language models (LLMs). From versions 0.10.2 to before 0.11.1, a memory corruption vulnerability could lead to a crash (denial-of-service) and potentially remote code execution (RCE), exists in the Completions API endpoint. When processing user-supplied prompt embeddings, the endpoint loads serialized tensors using torch.load() without sufficient validation. Due to a change introduced in PyTorch 2.8.0, sparse tensor integrity checks are disabled by default. As a result, maliciously crafted tensors can bypass internal bounds checks and trigger an out-of-bounds memory write during the call to to_dense(). This memory corruption can crash vLLM and potentially lead to code execution on the server hosting vLLM. This issue has been patched in version 0.11.1.

A vulnerability in vLLM allows attackers to supply malicious serialized prompt-embedding tensors that are deserialized using torch.load() without validation. Due to PyTorch 2.8.0 disabling sparse-tensor integrity checks by default, a crafted tensor can bypass bounds checks and cause an out-of-bounds write during to_dense(), leading to a crash (DoS) and potentially remote code execution on the vLLM server.