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., "repe
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.
왜 이 VPI인가 (설명가능 · 실험적)
VPI 산정 기준
| 영향도 | 65.00 |
| 악용 신호(추가 악용신호 없음) | ×1.00 |
| VPI | 65.00 |
VPI 공식 vpi-v1 기준
| 소스 | CVSS 버전 | 기본 점수 | 심각도 | 벡터 문자열 | 평가일 |
|---|---|---|---|---|---|
| NVDNIST | 3.1 | 6.5 | MEDIUM | CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H | 2026. 05. 13. |
| OSV3rd | 3.1 | 6.5 |
| CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H |
| 2026. 05. 13. |