CVE-2026-44223
Description
vLLM is an inference and serving engine for large language models (LLMs). From 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.
Predictions
Heuristic predictions, AS-IS, for prioritization only.
Mitigations
Vendor advisory: security-advisories@github.com — https://github.com/vllm-project/vllm/security/advisories/GHSA-83vm-p52w-f9pw
Vendor advisory: security-advisories@github.com — https://github.com/vllm-project/vllm/pull/38610
Application impact
| Vendor | Product | Versions | Fixed |
|---|---|---|---|
| vllm | vllm | {"startIncluding":"0.18.0","endExcluding":"0.20.0"} | 0.20.0 |
References
CWEs
CWE-131 CWE-704
Verify integrity in audit chain (admin only). AS-IS.