CVE-2026-44223
Fixed in a distro release EPSS <1%CVSS 6.5
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.
Products: vllm-project vllm
Fixed upstream
| Ecosystem | Package | Fixed in |
|---|---|---|
| PyPI | vllm | 0.20.0 |
Distro releases
| Release | Package | Fixed in |
|---|---|---|
| Chainguard | lmcache-cuda-12.8 | no fix |
| Chainguard | py3.10-vllm-cuda-12.4 | no fix |
| Chainguard | py3.10-vllm-cuda-12.9 | no fix |
| Chainguard | py3.12-vllm-cuda-12.4 | no fix |
| Chainguard | py3.12-vllm-cuda-12.9 | no fix |
| Chainguard | py3.13-vllm-cuda-12.9 | no fix |
| Chainguard | vllm-openai-cuda-12.9 | 0.28.0-r0 |
| Chainguard | vllm-openai-cuda-12.9-nvidia-cusparse | no fix |