{
  "apiVersion": "v1",
  "methodology": "https://sourcescore.org/methodology/",
  "canonical": "https://sourcescore.org/claims/468a9e2c047d8f2f/",
  "claim": {
    "vertical": "ai-ml",
    "subject": "vLLM",
    "predicate": "introduced_in",
    "object": "Kwon et al. 2023 — high-throughput LLM serving via PagedAttention",
    "confidence": 1,
    "sources": [
      {
        "url": "https://arxiv.org/abs/2309.06180",
        "title": "Efficient Memory Management for Large Language Model Serving with PagedAttention",
        "publisher": "arXiv (Kwon, Li, Zhuang, Sheng, Zheng, Yu, Gonzalez, Zhang, Stoica / UC Berkeley)",
        "publishedDate": "2023-09-12",
        "accessedDate": "2026-05-16",
        "type": "preprint",
        "excerpt": "We propose PagedAttention, an attention algorithm inspired by the classical virtual memory and paging techniques in operating systems. On top of it, we build vLLM, an LLM serving system that achieves (1) near-zero waste in KV cache memory and (2) flexible sharing of KV cache within and across requests to further reduce memory usage."
      },
      {
        "url": "https://github.com/vllm-project/vllm",
        "title": "vLLM — official GitHub repository",
        "publisher": "vLLM Project",
        "publishedDate": "2023-06-20",
        "accessedDate": "2026-05-16",
        "type": "github-release"
      }
    ],
    "publishedAt": "2026-05-16T00:00:00Z",
    "lastVerified": "2026-05-16",
    "methodologyVersion": "veritas-v0.1",
    "tags": [
      "vllm",
      "paged-attention",
      "uc-berkeley",
      "inference",
      "serving",
      "open-source",
      "2023",
      "introduced_in"
    ],
    "id": "468a9e2c047d8f2f",
    "statement": "vLLM introduced in: Kwon et al. 2023 — high-throughput LLM serving via PagedAttention."
  },
  "signature": {
    "algorithm": "HMAC-SHA256",
    "signedBy": "did:web:sourcescore.org",
    "signedAt": "2026-05-17T00:00:00.000Z",
    "signature": "e6114a4b1989988704c87888fcfe03df6b1bfea9dc38161d5ce5172d81c5b162"
  },
  "citedAs": "vLLM introduced in: Kwon et al. 2023 — high-throughput LLM serving via PagedAttention. — SourceScore Claim 468a9e2c047d8f2f (verified 2026-05-16, signed e6114a4b…). https://sourcescore.org/claims/468a9e2c047d8f2f/"
}