Signal is active with 1 mentions in latest observed window
Recommended action window: Monitor and triage in normal cycle
NVD description
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.
Priority
LOW
Exploitation
NONE
PoC
NONE
Patch
AVAILABLE
Momentum
NONE
If you run products in this scope, you should treat this CVE as relevant to your environment.
CVE-2026-44223 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 retu… https://www.cve.org/CVERecord?id=CVE-2026-44223
Post summary
A new vulnerability affecting vLLM’s extract_hidden_states speculative decoding proposer before version 0.20.0 has been documented; no PoC, exploit, or patch information is provided.
1 of 1 entries
| Part | Vendor | Product | Version | Target SW | Target HW |
|---|---|---|---|---|---|
| App | vllm | vllm | - | - | - |