Improper Input Validation Affecting vllm package, versions [0.5.5,0.18.0)
Threat Intelligence
The probability is the direct output of the EPSS model, and conveys an overall sense of the threat of exploitation in the wild. The percentile measures the EPSS probability relative to all known EPSS scores. Note: This data is updated daily, relying on the latest available EPSS model version. Check out the EPSS documentation for more details.
Do your applications use this vulnerable package?
In a few clicks we can analyze your entire application and see what components are vulnerable in your application, and suggest you quick fixes.
Test your applicationsSnyk Learn
Learn about Improper Input Validation vulnerabilities in an interactive lesson.
Start learning- Snyk IDSNYK-PYTHON-VLLM-15876757
- published3 Apr 2026
- disclosed2 Apr 2026
- creditkexinoh
Introduced: 2 Apr 2026
NewCVE-2026-34760 (opens in a new tab)How to fix?
Upgrade vllm to version 0.18.0 or higher.
Overview
vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs
Affected versions of this package are vulnerable to Improper Input Validation due to inconsistent downmixing behavior in the to_mono process. An attacker can manipulate audio inputs to cause the AI model to interpret audio differently from how it is perceived by humans, potentially leading to incorrect or malicious outputs by exploiting the discrepancy between standard audio playback and model processing.