p PatchFirst

CVE-2025-46560

CVE-2025-46560 · vLLM phi4mm: Quadratic Time Complexity in Input Token Processing​ leads to denial of service

This page contains facts from public vulnerability sources. Check the original advisory before updating.

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Versions starting from 0.8.0 and prior to 0.8.5 are affected by a critical performance vulnerability in the input preprocessing logic of the multimodal tokenizer. The code dynamically replaces placeholder tokens (e.g., <|audio_|>, <|image_|>) with repeated tokens based on precomputed lengths. Due to ​​inefficient list concatenation operations​​, the algorithm exhibits ​​quadratic time complexity (O(n²))​​, allowing malicious actors to trigger resource exhaustion via specially crafted inputs. This issue has been patched in version 0.8.5.

CVSS 6.5 · Medium · EPSS 0.5% · Fix available

Published
Updated

Stack: python

Affected products and versions

  • vllm

    Versions
    • >= 0.8.0, < 0.8.5: Affected

    No fix published

  • vllm PyPI

    Versions
    • 0.8.0
    • 0.8.1
    • 0.8.2
    • 0.8.3
    • 0.8.4
    Affected ranges
    • ECOSYSTEM: ≥ 0.8.0, < 0.8.5

    Update to 0.8.5

Original sources

Check your server →