Skip to content
OpenAdvisories
Medium WID-SEC-2026-3483 3 CVEs

vllm: Mehrere Schwachstellen ermöglichen Manipulation von Daten

Published Sep 20, 2026 · Updated Sep 27, 2026 · Revision 2 · de-DE

Vendor: SUSE Product: openSUSE

Summary

Ein Angreifer kann mehrere Schwachstellen in vllm ausnutzen, um Daten zu manipulieren.

Product

Open Source vLLM ist eine Open-Source-Bibliothek für schnelle und effiziente Inferenz von Large Language Models (LLMs).

Vulnerabilities 3

  • Low 3.7
    CVE-2026-93840

    vLLM before 0.29.0 Cross-Request Logits Corruption via allowed_token_ids

    vLLM before 0.29.0 validates allowed_token_ids against tokenizer length instead of model output logits width in SamplingParams._validate_allowed_token_ids(). Attackers can supply token IDs above the output vocabulary that pass validation, causing LogitBiasSta…

    vLLM vLLM vllm PyPI
  • Low 3.7
    CVE-2026-93841

    vLLM through 0.29.0 Adjacent-Request Sampler State Corruption via Unvalidated Prompt Token IDs

    vLLM through 0.29.0 contains a memory corruption vulnerability in the Triton _bincount_kernel where prompt token IDs index the penalty prompt-presence bitset without bounds checking against vocabulary size. Attackers can submit multimodal audio requests with …

    vLLM vLLM vllm PyPI
  • Low 3.1
    CVE-2026-93989

    vLLM through 0.29.0 Cross-Request Logits Corruption via bad_words

    vLLM through 0.29.0 fails to properly validate bad_words token indices against the model's generation output width in SamplingParams.update_from_tokenizer(). Attackers can supply out-of-bounds token indices that corrupt logits memory of concurrent requests, c…

    vLLM vLLM vllm PyPI

Revision history

  1. #2

    Neue Updates von openSUSE aufgenommen

  2. #1

    Initiale Fassung

References