CVE-2026-44223

Summary

vLLM is an inference and serving engine for large language models (LLMs). From 0.18.0 to before 0.20.0, the extract_hidden_states speculative decoding proposer in vLLM returns a tensor with an incorrect shape after the first decode step, causing a RuntimeError that crashes the EngineCore process. The crash is triggered when any request in the batch uses sampling penalty parameters (repetition_penalty, frequency_penalty, or presence_penalty). A single request with a penalty parameter (e.g., "repetition_penalty": 1.1) is sufficient to crash the server. This vulnerability is fixed in 0.20.0.

Affected Software

VendorProductVersion RangeStatus
vllm-projectvllm>= 0.18.0, < 0.20.0affected

Weaknesses

  • CWE-131: CWE-131: Incorrect Calculation of Buffer Size
  • CWE-704: CWE-704: Incorrect Type Conversion or Cast

ADP Enrichment

CISA ADP Vulnrichment

  • SSVC:
  • Exploitation: poc
    • Automatable: no
    • Technical Impact: partial

Additional References

References