vllm
Security Risk Profile
Security Risk Score
Comprehensive risk assessment based on 163 vulnerabilities, EPSS scores, exploitation status, and remediation availability.
📅 Data spans from September 12, 2024 to present
Threat Assessment
Severity Distribution
Exploit Likelihood
Age Distribution
Common Weaknesses (CWE)
Most Affected Products
Recent Vulnerabilities
See more →vLLM 0.14.0rc1 before 0.31.0 Heap Address Disclosure via sanitize_message()
vLLM 0.10.1 before 0.31.0 Cohere2 Vision DoS via max_patches
vLLM from 0.24.0 before 0.31.0 Resource Exhaustion via MiniMax M3 min_pixels
vLLM 0.28.0 before 0.30.0 Denial of Service via Prompt Embeds on XD-RoPE Models
vLLM 0.19.0 before 0.31.0 Resource Exhaustion via Request-Level num_frames
vLLM: Responses API bypasses trust_request_chat_template for chat_template_kwargs
vLLM: Permissive default CORS origins allow cross-origin requests to the vLLM HTTP server
vLLM: Out-of-vocabulary prompt token id passes input validation and triggers a CUDA device-side assert (engine crash / DoS)
vLLM: Untrusted EPS inputs can trigger denial of service or Ghostscript RCE
vLLM DeepSeek-OCR image processing vulnerable to denial of service
Monitor vllm in Real-Time
Get instant alerts when new vulnerabilities are discovered. Stay ahead of security threats with SecAlerts.