Skip to content
LLMHigh

CVE-2026-105758: vLLM is an inference and serving engine for large language models. From 0.24.0 until 0.30.0, the Qwe…

Occurred
Oct 5, 2026
Published
Oct 5, 2026
Source
NVD CVE

Summary

vLLM is an inference and serving engine for large language models. From 0.24.0 until 0.30.0, the Qwen2VLVideoBackend and Qwen3VLVideoBackend classes accept request-level values for the media_io_kwargs.video.max_frames and media_io_kwargs.video.fps fields without enforcing server-side ceilings. An unauthenticated caller can submit these values to the /tokenize endpoint, causing the sampler to decode every frame selected from attacker-controlled video input, consume disproportionate frontend memory, and potentially terminate the API process before scheduling or admission control. The Rust frontend is not affected because it rejects the media_io_kwargs field. This issue is fixed in version 0.30.0.

Tags

  • CVE
  • CVE-2026-105758
  • cvss:MEDIUM

Classification

Category confidence: 95%
Severity confidence: 90%

Rationale: CVE describes a vulnerability in vLLM LLM serving engine allowing resource abuse via unbounded frame processing.

Classifier: Qwen/Qwen3.5-35B-A3B-FP8

Visit source