What’s Vulnerable › Vulnerabilities

CVE-2026-44223: vLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parameters

• Severity: medium (CVSS 6.5 v3.1)
• Exploited in the wild: not listed in CISA KEV
• EPSS: 0.4% probability of exploitation in the next 30 days
• Public exploit or detection template: none known
• Published: 2026-05-12 · updated 2026-06-22
• Weakness: CWE-131

Description

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 products

• vllm-project vllm: from 0.18.0 before 0.20.0 — fixed in 0.20.0

How to fix

Upgrade to a fixed version: vllm-project vllm 0.20.0. Check the vendor advisory for mitigations.

References

• https://github.com/vllm-project/vllm/pull/38610
• https://github.com/vllm-project/vllm/security/advisories/GHSA-83vm-p52w-f9pw
• https://github.com/vllm-project/vllm/pull/38610
• https://github.com/vllm-project/vllm/security/advisories/GHSA-83vm-p52w-f9pw

Get alerted when CVE-2026-44223 or the next vulnerability affects your stack