← Database

CVE-2026-116180

HIGH CVSS: 7.7 Published: 2026-10-25 Modified: 2026-10-25

Description

vLLM from 0.8.5 before 0.31.0 has a denial of service flaw. LogprobsTensors.empty_cpu() in vllm/v1/outputs.py allocates an uninitialized int32 tensor, and the V1 model runner leaves prompt positions reported as computed by a KV connector unwritten. A client sharing the engine can send prompt_logprobs (or echo with logprobs) against a connector-cached prefix, making tokenizer decode raise OverflowError, which crashes EngineCore and drops all co-tenant requests.

Affected Platforms

Unknown

Weakness Type

CWE-908

References