2 matches found
Allocation of Resources Without Limits or Throttling
Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling in the loadaudiopyav. An attacker can exhaust system memory by submitting a small, highly...
7.5CVSS5.8AI score0.00201EPSS
SaveExploits0References2
CVE-2026-90554
vLLM versions =0.10.2 and 0.28.0 do not apply any audio decode-size or duration limit when extracting audio from video input for NanoNemotronVL models. In nanonemotronvl.py, extractaudiofromvideos calls loadaudiopyavBytesIOvideobytes without the maxdurations or maxdecodebytes parameters, so neith...
7.5CVSS5.8AI score0.00201EPSS
SaveExploits0References4
20