medium6.5
CVE-2026-44223
vLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parameters
Impact
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 packages
| Ecosystem | Package | Affected | Fixed in |
|---|---|---|---|
| pip | vllm | >= 0.18.0, < 0.20.0 | 0.20.0 |
Affected products
| Vendor | Product | Affected versions |
|---|---|---|
| Vllm | Vllm | from 0.18.0 before 0.20.0 |
Weakness classification
References
- NVD record
- CVE.org record
- GitHub advisory
- https://github.com/advisories/GHSA-83vm-p52w-f9pw
- https://github.com/vllm-project/vllm/security/advisories/GHSA-83vm-p52w-f9pw
- https://github.com/vllm-project/vllm/pull/38610
- https://nvd.nist.gov/vuln/detail/CVE-2026-44223
- https://github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2026-145.yaml