{
 "id": "AIR-2026-0213",
 "slug": "vllm-decoder-allocation-resource-limit-bypass-causes-gpu-mem",
 "title": "vLLM Decoder Allocation Resource Limit Bypass Causes GPU Memory Exhaustion",
 "summary": "vLLM before 0.29.0 contains a resource-limit bypass in PyNvVideoCodec decoder allocation. Attackers can select specific sampler subclasses in video requests to exceed decoder limits and exhaust GPU memory.",
 "type": "attack_on_ai",
 "categories": [
  "infra_vuln"
 ],
 "severity": 2,
 "targets": [
  "framework_or_tooling"
 ],
 "products": [
  "vLLM"
 ],
 "incident_date": "2026-09-26",
 "incident_date_precision": "day",
 "first_reported": "2026-09-26",
 "sources": [
  {
   "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-100649",
   "publisher": "NVD",
   "title": "CVE-2026-100649: vLLM before 0.29.0 contains a resource-limit bypass vulnerability in PyNvVideoCodec decoder allocation where sampler subclass…",
   "published_at": "2026-09-26"
  }
 ],
 "source_count": 1,
 "confidence": 0.95,
 "status": "published",
 "origin": "seed_nvd",
 "tags": [
  "CVE-2026-100649",
  "resource_exhaustion",
  "gpu_memory"
 ],
 "created_at": "2026-10-18T06:33:31.967Z",
 "updated_at": "2026-10-18T06:33:31.967Z"
}