{
  "schema_version": 1,
  "category": "Vector databases",
  "slug": "vector-databases",
  "niche": "devops",
  "niche_name": "DevOps & developer-infrastructure tools",
  "niche_short": "DevOps & dev tools",
  "description": "Which vector databases do AI engines recommend when developers building RAG and AI features ask what to use to store and search embeddings?",
  "audience": "AI/ML and application engineers, and the founders + DevRel who sell vector search infrastructure",
  "methodology": {
    "metric": "share-of-model",
    "metric_definition": "Share-of-model = the percentage of all product recommendations, across every buyer prompt and engine, that name a given product. It answers: when an AI recommends something in this category, how often is it this product?",
    "runs_per_prompt_per_engine": 10,
    "engines": [
      "Perplexity",
      "Google Gemini",
      "ChatGPT (OpenAI)",
      "Claude (Anthropic)",
      "Grok (xAI)"
    ],
    "engine_keys": [
      "perplexity",
      "gemini",
      "openai",
      "anthropic",
      "grok"
    ],
    "engines_measured": [
      "Perplexity",
      "Google Gemini",
      "ChatGPT (OpenAI)",
      "Claude (Anthropic)",
      "Grok (xAI)"
    ],
    "engine_keys_measured": [
      "perplexity",
      "gemini",
      "openai",
      "anthropic",
      "grok"
    ],
    "min_n_per_engine": 5,
    "coverage_note": "",
    "n_prompts": 10,
    "total_answers": 500,
    "errors": 0,
    "confidence_interval": "Wilson score interval, 95% (z=1.96)",
    "sample_size_note": "n=500: this supports a ±5pp claim (~±4pp at this n; ~385 answers needed for ±5pp).",
    "provenance": {
      "surface": "api",
      "mode": "live",
      "region": "US/English",
      "engines": [
        {
          "engine": "perplexity",
          "display": "Perplexity",
          "model": "sonar",
          "api_version": "api.perplexity.ai/chat/completions",
          "grounding": true
        },
        {
          "engine": "gemini",
          "display": "Gemini",
          "model": "gemini-2.5-flash",
          "api_version": "generativelanguage/v1beta:generateContent",
          "grounding": true
        },
        {
          "engine": "openai",
          "display": "ChatGPT",
          "model": "gpt-4.1",
          "api_version": "api.openai.com/v1/responses",
          "grounding": true
        },
        {
          "engine": "anthropic",
          "display": "Claude",
          "model": "claude-sonnet-4-5",
          "api_version": "api.anthropic.com/v1/messages (2023-06-01)",
          "grounding": true
        },
        {
          "engine": "grok",
          "display": "Grok",
          "model": "grok-3",
          "api_version": "api.x.ai/v1/chat/completions",
          "grounding": true
        }
      ],
      "grounding_all_on": true,
      "approximation_note": "API answers approximate, but do not exactly replicate, what a human sees in the consumer app (different system prompts, tools, and browsing defaults).",
      "consumer_surfaces": [
        {
          "key": "google_ai_overview",
          "label": "Google AI Overview",
          "via": "dataforseo",
          "grounding": true,
          "note": "Zero-click AI answer block on Google SERPs; no official API — measured via the DataForSEO AI Mode SERP. A clicked citation passes to analytics as google/organic, so referral tracking under-counts it (see GSC's Generative-AI report)."
        },
        {
          "key": "google_ai_mode",
          "label": "Google AI Mode",
          "via": "dataforseo",
          "grounding": true,
          "note": "Google's conversational AI search; no official API — measured via DataForSEO."
        },
        {
          "key": "chatgpt_consumer",
          "label": "ChatGPT (consumer app)",
          "via": "dataforseo",
          "grounding": true,
          "note": "The consumer ChatGPT app (system prompt, tools, browsing defaults) differs from the API; measured via the DataForSEO LLM scraper, not our API adapter."
        },
        {
          "key": "bing_copilot_answers",
          "label": "Bing's Copilot answers",
          "via": "dataforseo",
          "grounding": true,
          "note": "Microsoft's AI answer block on Bing search results (Copilot Search); a separate search surface from the standalone Microsoft Copilot engine — no official public API, measured via the DataForSEO Bing SERP ai_overview element (the same modality already used for Google AI Overview)."
        },
        {
          "key": "copilot_workiq",
          "label": "Microsoft Copilot (Work IQ)",
          "via": "workiq",
          "grounding": true,
          "note": "Microsoft 365 Copilot's synthesized answer + citations via the Work IQ Chat API. TENANT-SCOPED and therefore DISQUALIFIED as a brand-visibility measurement source (X43) - named here so the surface stays describable, never measured."
        }
      ],
      "captured_at_utc": "2026-08-07T23:50:30+00:00"
    },
    "detection": "A product is counted as 'recommended' when its name or a known alias is mentioned (case-insensitive, word-boundary aware) in the answer. Citations are counted when the product's own domain appears in the answer's source links. Detection is heuristic — a mention is not always a positive endorsement.",
    "mode": "live",
    "is_illustrative": false
  },
  "pre_registration": {
    "prompts_hash": "57ac86b787ad0242",
    "n_prompts": 10,
    "brand_universe": [
      "Chroma",
      "LanceDB",
      "Milvus",
      "Pinecone",
      "Qdrant",
      "Redis",
      "Turbopuffer",
      "Vespa",
      "Weaviate",
      "pgvector"
    ],
    "brand_universe_hash": "4930306d8f150a5a",
    "n_brands": 10,
    "metric": "share-of-model",
    "metrics_declared": [
      "share_of_model",
      "appearance_rate",
      "citation_rate"
    ],
    "frozen_before_run": true,
    "mode": "live",
    "note": "Prompt set + brand universe + metrics were pre-registered (fixed) before this run; the published leaderboard is drawn only from this universe — no post-hoc cherry-picking."
  },
  "coi": {
    "policy": "exclude_own_clients",
    "applies": true,
    "client_domains_checked": 0,
    "excluded": [],
    "disclosure": "Clear Cited excludes its own clients from this public ranked Index to avoid a conflict of interest. Excluded brands may still be measured, but are not ranked here; the policy is disclosed rather than the exclusion hidden."
  },
  "license": {
    "name": "CC BY 4.0",
    "url": "https://creativecommons.org/licenses/by/4.0/"
  },
  "prompts": [
    "What's the best vector database for a Series A startup building RAG in 2026?",
    "Best vector database for a small team adding semantic search?",
    "What vector database should we use for a Kubernetes microservices stack?",
    "Most cost-effective vector database for a high-traffic AI app?",
    "Best vector database for a team already on Postgres?",
    "Pinecone vs Weaviate for a production RAG pipeline?",
    "Qdrant vs Milvus for self-hosted vector search?",
    "What's a good Pinecone alternative that is open-source?",
    "Best vector database for hybrid keyword + semantic search?",
    "Which vector database do AI engineering teams recommend for scale?"
  ],
  "products_field": [
    {
      "name": "Pinecone",
      "domain": "pinecone.io"
    },
    {
      "name": "Weaviate",
      "domain": "weaviate.io"
    },
    {
      "name": "Qdrant",
      "domain": "qdrant.tech"
    },
    {
      "name": "Chroma",
      "domain": "trychroma.com"
    },
    {
      "name": "Milvus",
      "domain": "milvus.io"
    },
    {
      "name": "pgvector",
      "domain": "github.com"
    },
    {
      "name": "Redis",
      "domain": "redis.io"
    },
    {
      "name": "Vespa",
      "domain": "vespa.ai"
    },
    {
      "name": "LanceDB",
      "domain": "lancedb.com"
    },
    {
      "name": "Turbopuffer",
      "domain": "turbopuffer.com"
    }
  ],
  "run_date": "2026-08-07",
  "last_updated": "2026-08-07",
  "leaderboard": [
    {
      "product": "Qdrant",
      "domain": "qdrant.tech",
      "mentions": 432,
      "share_of_model": 0.1969,
      "share_ci": [
        0.1808,
        0.2141
      ],
      "appearance_rate": 0.864,
      "appearance_ci": [
        0.8312,
        0.8913
      ],
      "citation_rate": 0.014,
      "citations": 7,
      "by_engine": {
        "perplexity": {
          "appearance_rate": 0.9,
          "appearance_ci": [
            0.8256,
            0.9448
          ],
          "n": 100,
          "mentions": 90
        },
        "gemini": {
          "appearance_rate": 0.87,
          "appearance_ci": [
            0.7902,
            0.9224
          ],
          "n": 100,
          "mentions": 87
        },
        "openai": {
          "appearance_rate": 0.82,
          "appearance_ci": [
            0.7333,
            0.883
          ],
          "n": 100,
          "mentions": 82
        },
        "anthropic": {
          "appearance_rate": 0.88,
          "appearance_ci": [
            0.8019,
            0.93
          ],
          "n": 100,
          "mentions": 88
        },
        "grok": {
          "appearance_rate": 0.85,
          "appearance_ci": [
            0.7672,
            0.9069
          ],
          "n": 100,
          "mentions": 85
        }
      },
      "rank": 1
    },
    {
      "product": "Weaviate",
      "domain": "weaviate.io",
      "mentions": 426,
      "share_of_model": 0.1942,
      "share_ci": [
        0.1782,
        0.2112
      ],
      "appearance_rate": 0.852,
      "appearance_ci": [
        0.8182,
        0.8804
      ],
      "citation_rate": 0.024,
      "citations": 12,
      "by_engine": {
        "perplexity": {
          "appearance_rate": 0.81,
          "appearance_ci": [
            0.7222,
            0.8749
          ],
          "n": 100,
          "mentions": 81
        },
        "gemini": {
          "appearance_rate": 0.87,
          "appearance_ci": [
            0.7902,
            0.9224
          ],
          "n": 100,
          "mentions": 87
        },
        "openai": {
          "appearance_rate": 0.91,
          "appearance_ci": [
            0.8377,
            0.9519
          ],
          "n": 100,
          "mentions": 91
        },
        "anthropic": {
          "appearance_rate": 0.79,
          "appearance_ci": [
            0.7002,
            0.8583
          ],
          "n": 100,
          "mentions": 79
        },
        "grok": {
          "appearance_rate": 0.88,
          "appearance_ci": [
            0.8019,
            0.93
          ],
          "n": 100,
          "mentions": 88
        }
      },
      "rank": 2
    },
    {
      "product": "Pinecone",
      "domain": "pinecone.io",
      "mentions": 419,
      "share_of_model": 0.191,
      "share_ci": [
        0.1751,
        0.208
      ],
      "appearance_rate": 0.838,
      "appearance_ci": [
        0.8031,
        0.8677
      ],
      "citation_rate": 0.042,
      "citations": 21,
      "by_engine": {
        "perplexity": {
          "appearance_rate": 0.77,
          "appearance_ci": [
            0.6785,
            0.8416
          ],
          "n": 100,
          "mentions": 77
        },
        "gemini": {
          "appearance_rate": 0.85,
          "appearance_ci": [
            0.7672,
            0.9069
          ],
          "n": 100,
          "mentions": 85
        },
        "openai": {
          "appearance_rate": 0.88,
          "appearance_ci": [
            0.8019,
            0.93
          ],
          "n": 100,
          "mentions": 88
        },
        "anthropic": {
          "appearance_rate": 0.81,
          "appearance_ci": [
            0.7222,
            0.8749
          ],
          "n": 100,
          "mentions": 81
        },
        "grok": {
          "appearance_rate": 0.88,
          "appearance_ci": [
            0.8019,
            0.93
          ],
          "n": 100,
          "mentions": 88
        }
      },
      "rank": 3
    },
    {
      "product": "Milvus",
      "domain": "milvus.io",
      "mentions": 335,
      "share_of_model": 0.1527,
      "share_ci": [
        0.1382,
        0.1683
      ],
      "appearance_rate": 0.67,
      "appearance_ci": [
        0.6276,
        0.7098
      ],
      "citation_rate": 0.04,
      "citations": 20,
      "by_engine": {
        "perplexity": {
          "appearance_rate": 0.57,
          "appearance_ci": [
            0.4722,
            0.6627
          ],
          "n": 100,
          "mentions": 57
        },
        "gemini": {
          "appearance_rate": 0.73,
          "appearance_ci": [
            0.6357,
            0.8073
          ],
          "n": 100,
          "mentions": 73
        },
        "openai": {
          "appearance_rate": 0.67,
          "appearance_ci": [
            0.5731,
            0.7544
          ],
          "n": 100,
          "mentions": 67
        },
        "anthropic": {
          "appearance_rate": 0.59,
          "appearance_ci": [
            0.492,
            0.6813
          ],
          "n": 100,
          "mentions": 59
        },
        "grok": {
          "appearance_rate": 0.79,
          "appearance_ci": [
            0.7002,
            0.8583
          ],
          "n": 100,
          "mentions": 79
        }
      },
      "rank": 4
    },
    {
      "product": "pgvector",
      "domain": "github.com",
      "mentions": 323,
      "share_of_model": 0.1472,
      "share_ci": [
        0.133,
        0.1627
      ],
      "appearance_rate": 0.646,
      "appearance_ci": [
        0.6031,
        0.6867
      ],
      "citation_rate": 0.026,
      "citations": 13,
      "by_engine": {
        "perplexity": {
          "appearance_rate": 0.68,
          "appearance_ci": [
            0.5834,
            0.7633
          ],
          "n": 100,
          "mentions": 68
        },
        "gemini": {
          "appearance_rate": 0.67,
          "appearance_ci": [
            0.5731,
            0.7544
          ],
          "n": 100,
          "mentions": 67
        },
        "openai": {
          "appearance_rate": 0.49,
          "appearance_ci": [
            0.3942,
            0.5865
          ],
          "n": 100,
          "mentions": 49
        },
        "anthropic": {
          "appearance_rate": 0.69,
          "appearance_ci": [
            0.5937,
            0.7722
          ],
          "n": 100,
          "mentions": 69
        },
        "grok": {
          "appearance_rate": 0.7,
          "appearance_ci": [
            0.6041,
            0.7811
          ],
          "n": 100,
          "mentions": 70
        }
      },
      "rank": 5
    },
    {
      "product": "Chroma",
      "domain": "trychroma.com",
      "mentions": 142,
      "share_of_model": 0.0647,
      "share_ci": [
        0.0552,
        0.0758
      ],
      "appearance_rate": 0.284,
      "appearance_ci": [
        0.2462,
        0.3251
      ],
      "citation_rate": 0.0,
      "citations": 0,
      "by_engine": {
        "perplexity": {
          "appearance_rate": 0.13,
          "appearance_ci": [
            0.0776,
            0.2098
          ],
          "n": 100,
          "mentions": 13
        },
        "gemini": {
          "appearance_rate": 0.34,
          "appearance_ci": [
            0.2546,
            0.4372
          ],
          "n": 100,
          "mentions": 34
        },
        "openai": {
          "appearance_rate": 0.32,
          "appearance_ci": [
            0.2367,
            0.4166
          ],
          "n": 100,
          "mentions": 32
        },
        "anthropic": {
          "appearance_rate": 0.23,
          "appearance_ci": [
            0.1584,
            0.3215
          ],
          "n": 100,
          "mentions": 23
        },
        "grok": {
          "appearance_rate": 0.4,
          "appearance_ci": [
            0.3094,
            0.498
          ],
          "n": 100,
          "mentions": 40
        }
      },
      "rank": 6
    },
    {
      "product": "Redis",
      "domain": "redis.io",
      "mentions": 37,
      "share_of_model": 0.0169,
      "share_ci": [
        0.0123,
        0.0232
      ],
      "appearance_rate": 0.074,
      "appearance_ci": [
        0.0542,
        0.1003
      ],
      "citation_rate": 0.036,
      "citations": 18,
      "by_engine": {
        "perplexity": {
          "appearance_rate": 0.0,
          "appearance_ci": [
            0.0,
            0.037
          ],
          "n": 100,
          "mentions": 0
        },
        "gemini": {
          "appearance_rate": 0.25,
          "appearance_ci": [
            0.1755,
            0.343
          ],
          "n": 100,
          "mentions": 25
        },
        "openai": {
          "appearance_rate": 0.02,
          "appearance_ci": [
            0.0055,
            0.07
          ],
          "n": 100,
          "mentions": 2
        },
        "anthropic": {
          "appearance_rate": 0.08,
          "appearance_ci": [
            0.0411,
            0.15
          ],
          "n": 100,
          "mentions": 8
        },
        "grok": {
          "appearance_rate": 0.02,
          "appearance_ci": [
            0.0055,
            0.07
          ],
          "n": 100,
          "mentions": 2
        }
      },
      "rank": 7
    },
    {
      "product": "Turbopuffer",
      "domain": "turbopuffer.com",
      "mentions": 28,
      "share_of_model": 0.0128,
      "share_ci": [
        0.0088,
        0.0184
      ],
      "appearance_rate": 0.056,
      "appearance_ci": [
        0.039,
        0.0797
      ],
      "citation_rate": 0.0,
      "citations": 0,
      "by_engine": {
        "perplexity": {
          "appearance_rate": 0.06,
          "appearance_ci": [
            0.0278,
            0.1248
          ],
          "n": 100,
          "mentions": 6
        },
        "gemini": {
          "appearance_rate": 0.1,
          "appearance_ci": [
            0.0552,
            0.1744
          ],
          "n": 100,
          "mentions": 10
        },
        "openai": {
          "appearance_rate": 0.0,
          "appearance_ci": [
            0.0,
            0.037
          ],
          "n": 100,
          "mentions": 0
        },
        "anthropic": {
          "appearance_rate": 0.11,
          "appearance_ci": [
            0.0625,
            0.1863
          ],
          "n": 100,
          "mentions": 11
        },
        "grok": {
          "appearance_rate": 0.01,
          "appearance_ci": [
            0.0018,
            0.0545
          ],
          "n": 100,
          "mentions": 1
        }
      },
      "rank": 8
    },
    {
      "product": "Vespa",
      "domain": "vespa.ai",
      "mentions": 27,
      "share_of_model": 0.0123,
      "share_ci": [
        0.0085,
        0.0178
      ],
      "appearance_rate": 0.054,
      "appearance_ci": [
        0.0374,
        0.0774
      ],
      "citation_rate": 0.0,
      "citations": 0,
      "by_engine": {
        "perplexity": {
          "appearance_rate": 0.0,
          "appearance_ci": [
            0.0,
            0.037
          ],
          "n": 100,
          "mentions": 0
        },
        "gemini": {
          "appearance_rate": 0.1,
          "appearance_ci": [
            0.0552,
            0.1744
          ],
          "n": 100,
          "mentions": 10
        },
        "openai": {
          "appearance_rate": 0.06,
          "appearance_ci": [
            0.0278,
            0.1248
          ],
          "n": 100,
          "mentions": 6
        },
        "anthropic": {
          "appearance_rate": 0.01,
          "appearance_ci": [
            0.0018,
            0.0545
          ],
          "n": 100,
          "mentions": 1
        },
        "grok": {
          "appearance_rate": 0.1,
          "appearance_ci": [
            0.0552,
            0.1744
          ],
          "n": 100,
          "mentions": 10
        }
      },
      "rank": 9
    },
    {
      "product": "LanceDB",
      "domain": "lancedb.com",
      "mentions": 25,
      "share_of_model": 0.0114,
      "share_ci": [
        0.0077,
        0.0168
      ],
      "appearance_rate": 0.05,
      "appearance_ci": [
        0.0341,
        0.0728
      ],
      "citation_rate": 0.0,
      "citations": 0,
      "by_engine": {
        "perplexity": {
          "appearance_rate": 0.0,
          "appearance_ci": [
            0.0,
            0.037
          ],
          "n": 100,
          "mentions": 0
        },
        "gemini": {
          "appearance_rate": 0.05,
          "appearance_ci": [
            0.0215,
            0.1118
          ],
          "n": 100,
          "mentions": 5
        },
        "openai": {
          "appearance_rate": 0.0,
          "appearance_ci": [
            0.0,
            0.037
          ],
          "n": 100,
          "mentions": 0
        },
        "anthropic": {
          "appearance_rate": 0.04,
          "appearance_ci": [
            0.0157,
            0.0984
          ],
          "n": 100,
          "mentions": 4
        },
        "grok": {
          "appearance_rate": 0.16,
          "appearance_ci": [
            0.101,
            0.2442
          ],
          "n": 100,
          "mentions": 16
        }
      },
      "rank": 10
    }
  ],
  "citation_sources": {
    "total_citations": 4539,
    "third_party_citations": 4447,
    "off_site_pct": 0.9797,
    "top": [
      {
        "domain": "vertexaisearch.cloud.google.com",
        "count": 1504,
        "share_of_citations": 0.3314,
        "first_party": false,
        "type": "other"
      },
      {
        "domain": "medium.com",
        "count": 202,
        "share_of_citations": 0.0445,
        "first_party": false,
        "type": "community"
      },
      {
        "domain": "firecrawl.dev",
        "count": 136,
        "share_of_citations": 0.03,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "iternal.ai",
        "count": 80,
        "share_of_citations": 0.0176,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "encore.dev",
        "count": 77,
        "share_of_citations": 0.017,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "reddit.com",
        "count": 72,
        "share_of_citations": 0.0159,
        "first_party": false,
        "type": "community"
      },
      {
        "domain": "zenml.io",
        "count": 72,
        "share_of_citations": 0.0159,
        "first_party": false,
        "type": "other"
      },
      {
        "domain": "alphacorp.ai",
        "count": 68,
        "share_of_citations": 0.015,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "kunalganglani.com",
        "count": 62,
        "share_of_citations": 0.0137,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "groovyweb.co",
        "count": 57,
        "share_of_citations": 0.0126,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "marktechpost.com",
        "count": 55,
        "share_of_citations": 0.0121,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "dev.to",
        "count": 54,
        "share_of_citations": 0.0119,
        "first_party": false,
        "type": "community"
      },
      {
        "domain": "myengineeringpath.dev",
        "count": 50,
        "share_of_citations": 0.011,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "atlan.com",
        "count": 47,
        "share_of_citations": 0.0104,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "pecollective.com",
        "count": 42,
        "share_of_citations": 0.0093,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "linkedin.com",
        "count": 40,
        "share_of_citations": 0.0088,
        "first_party": false,
        "type": "other"
      },
      {
        "domain": "youtube.com",
        "count": 40,
        "share_of_citations": 0.0088,
        "first_party": false,
        "type": "other"
      },
      {
        "domain": "techsy.io",
        "count": 37,
        "share_of_citations": 0.0082,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "metacto.com",
        "count": 34,
        "share_of_citations": 0.0075,
        "first_party": false,
        "type": "comparison"
      },
      {
        "domain": "shaped.ai",
        "count": 33,
        "share_of_citations": 0.0073,
        "first_party": false,
        "type": "comparison"
      }
    ]
  },
  "research": {
    "source_types": {
      "review": 20,
      "community": 339,
      "news": 0,
      "docs": 40,
      "comparison": 1845,
      "other": 2203
    },
    "winning_content_types": [
      {
        "type": "other",
        "citations": 2203
      },
      {
        "type": "comparison",
        "citations": 1845
      },
      {
        "type": "community",
        "citations": 339
      },
      {
        "type": "docs",
        "citations": 40
      },
      {
        "type": "review",
        "citations": 20
      }
    ],
    "referral_sites": {
      "review": [],
      "community": [
        "medium.com",
        "reddit.com",
        "dev.to"
      ],
      "news": [],
      "docs": [],
      "comparison": [
        "firecrawl.dev",
        "iternal.ai",
        "encore.dev",
        "alphacorp.ai",
        "kunalganglani.com"
      ],
      "other": [
        "vertexaisearch.cloud.google.com",
        "zenml.io",
        "linkedin.com",
        "youtube.com"
      ]
    },
    "engine_top_sources": {
      "perplexity": [
        {
          "domain": "medium.com",
          "count": 107,
          "type": "community"
        },
        {
          "domain": "reddit.com",
          "count": 72,
          "type": "community"
        },
        {
          "domain": "firecrawl.dev",
          "count": 71,
          "type": "comparison"
        },
        {
          "domain": "zenml.io",
          "count": 71,
          "type": "other"
        },
        {
          "domain": "encore.dev",
          "count": 70,
          "type": "comparison"
        }
      ],
      "gemini": [
        {
          "domain": "vertexaisearch.cloud.google.com",
          "count": 1504,
          "type": "other"
        }
      ],
      "openai": [
        {
          "domain": "appstackbuilder.com",
          "count": 31,
          "type": "comparison"
        },
        {
          "domain": "vectorwiki.com",
          "count": 21,
          "type": "comparison"
        },
        {
          "domain": "finance.yahoo.com",
          "count": 20,
          "type": "other"
        },
        {
          "domain": "genaizurich.ch",
          "count": 17,
          "type": "other"
        },
        {
          "domain": "inferensys.com",
          "count": 16,
          "type": "comparison"
        }
      ],
      "anthropic": [
        {
          "domain": "medium.com",
          "count": 94,
          "type": "community"
        },
        {
          "domain": "firecrawl.dev",
          "count": 65,
          "type": "comparison"
        },
        {
          "domain": "alphacorp.ai",
          "count": 48,
          "type": "comparison"
        },
        {
          "domain": "groovyweb.co",
          "count": 41,
          "type": "comparison"
        },
        {
          "domain": "pecollective.com",
          "count": 41,
          "type": "comparison"
        }
      ],
      "grok": []
    },
    "engine_citation_overlap_mean": 0.066,
    "engine_disagreement_note": "Mean Jaccard overlap of cited-source domains across engine pairs; lower = the engines cite different sources (they disagree on whom to trust).",
    "prompts": [
      "What's the best vector database for a Series A startup building RAG in 2026?",
      "Best vector database for a small team adding semantic search?",
      "What vector database should we use for a Kubernetes microservices stack?",
      "Most cost-effective vector database for a high-traffic AI app?",
      "Best vector database for a team already on Postgres?",
      "Pinecone vs Weaviate for a production RAG pipeline?",
      "Qdrant vs Milvus for self-hosted vector search?",
      "What's a good Pinecone alternative that is open-source?",
      "Best vector database for hybrid keyword + semantic search?",
      "Which vector database do AI engineering teams recommend for scale?"
    ],
    "n_prompts": 10,
    "engines": [
      "perplexity",
      "gemini",
      "openai",
      "anthropic",
      "grok"
    ]
  },
  "leader_insight": [
    {
      "product": "Qdrant",
      "domain": "qdrant.tech",
      "rank": 1,
      "share_pct": 19.7,
      "share_ci_pct": [
        18.1,
        21.4
      ],
      "appearance_pct": 86.4,
      "citation_rate_pct": 1.4,
      "strongest_engine": "Perplexity",
      "lead_drivers": [
        "highest share-of-model in the field",
        "strongest on Perplexity"
      ],
      "category_source_types": {
        "review": 20,
        "community": 339,
        "news": 0,
        "docs": 40,
        "comparison": 1845,
        "other": 2203
      },
      "gaps": [
        "low citation rate — their own domain is rarely the cited source"
      ]
    }
  ],
  "headline_stats": [
    "Qdrant leads our AI Visibility Index for Vector databases at 20% share-of-model across 5 AI engines.",
    "When AI recommends Vector databases, the top 3 by share-of-model are Qdrant, Weaviate, Pinecone — measured over 10+ runs per engine, not a single screenshot.",
    "Even the Vector databases leader, Qdrant, swings from 90% on Perplexity to 82% on ChatGPT (OpenAI). AI visibility is engine-specific."
  ],
  "headline_stats_tagged": [
    {
      "niche": "devops",
      "niche_short": "DevOps & dev tools",
      "category": "Vector databases",
      "slug": "vector-databases",
      "text": "Qdrant leads our AI Visibility Index for Vector databases at 20% share-of-model across 5 AI engines.",
      "hashtags": [
        "#DevOps",
        "#DevTools",
        "#SRE",
        "#PlatformEngineering",
        "#Observability",
        "#Kubernetes"
      ],
      "url": "https://clearcited.com/ai-visibility-index/vector-databases/"
    },
    {
      "niche": "devops",
      "niche_short": "DevOps & dev tools",
      "category": "Vector databases",
      "slug": "vector-databases",
      "text": "When AI recommends Vector databases, the top 3 by share-of-model are Qdrant, Weaviate, Pinecone — measured over 10+ runs per engine, not a single screenshot.",
      "hashtags": [
        "#DevOps",
        "#DevTools",
        "#SRE",
        "#PlatformEngineering",
        "#Observability",
        "#Kubernetes"
      ],
      "url": "https://clearcited.com/ai-visibility-index/vector-databases/"
    },
    {
      "niche": "devops",
      "niche_short": "DevOps & dev tools",
      "category": "Vector databases",
      "slug": "vector-databases",
      "text": "Even the Vector databases leader, Qdrant, swings from 90% on Perplexity to 82% on ChatGPT (OpenAI). AI visibility is engine-specific.",
      "hashtags": [
        "#DevOps",
        "#DevTools",
        "#SRE",
        "#PlatformEngineering",
        "#Observability",
        "#Kubernetes"
      ],
      "url": "https://clearcited.com/ai-visibility-index/vector-databases/"
    }
  ],
  "provider": {
    "name": "Clear Cited",
    "url": "https://clearcited.com",
    "index_url": "https://clearcited.com/ai-visibility-index/"
  },
  "disclaimer": "Measurements reflect a point in time; AI engines change continuously. API answers approximate, but do not exactly replicate, the consumer apps. Clear Cited does not guarantee any product's ranking. "
}