CVE-2026-44223GHSA-83VM-P52W-F9PWPYSEC-2026-145
vLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parameters
- Published
- Last modified
- Where the record stands: published, rejected by its publisher, or superseded by another id.What do these mean? →
- published
- Whether CISA lists it in its Known Exploited Vulnerabilities catalog: proof that attackers are using it.What do these mean? →
- Not listed in CISA KEV
- Fix
- Fixed in 0.20.0
- Whether working attack code is public (Exploit-DB, Metasploit, Nuclei or GitHub). Public code makes attacks easier.What do these mean? →
- None on Exploit-DB, Metasploit or GitHub
CISA's decision points: whether it is exploited, whether attacks can be automated, and how much control an attacker gets. Assessed (SSVC 2.0.3).What do these mean? →
- Exploitation: poc
- Automatable: no
- Technical impact: partial
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.
Timeline
- CVE published
- CISA SSVC assessment
Exploitation: poc · Automatable: no · Technical impact: partial
- Record last updated
Affected versions
- vllm-project vllm
From 0.18.0 · fixed in 0.20.0
- vllmPyPI
From 0.18.0 · fixed in 0.20.0