{
 "id": "AIR-2026-0078",
 "slug": "cve-2026-73557-vllm-prompt-embed-check-bypass-vulnerability",
 "title": "CVE-2026-73557: vLLM prompt embed check bypass vulnerability",
 "summary": "A race condition in vLLM allows concurrent prompt embeds to bypass sparse tensor checks and reach the dense tensor conversion. The issue was addressed in version 0.26.0.",
 "type": "attack_on_ai",
 "categories": [
  "infra_vuln"
 ],
 "severity": 2,
 "targets": [
  "framework_or_tooling"
 ],
 "products": [
  "vLLM"
 ],
 "incident_date": "2026-08-13",
 "incident_date_precision": "day",
 "first_reported": "2026-08-13",
 "sources": [
  {
   "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-73557",
   "publisher": "NVD",
   "title": "CVE-2026-73557: vLLM is an inference and serving engine for large language models.",
   "published_at": "2026-08-13"
  },
  {
   "url": "https://github.com/advisories/GHSA-pr7f-p5mw-fc87",
   "publisher": "GitHub Advisory Database",
   "title": "CVE-2026-73557:  vLLM: Incomplete CVE-2025-62164 remediation can be bypassed by concurrent prompt parts",
   "published_at": "2026-09-04"
  }
 ],
 "source_count": 2,
 "confidence": 0.95,
 "status": "published",
 "origin": "seed_nvd",
 "tags": [
  "CVE-2026-73557",
  "vllm"
 ],
 "created_at": "2026-10-18T06:33:31.967Z",
 "updated_at": "2026-10-18T18:20:34.882Z"
}