{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Organization",
      "@id": "https://onpremisai.be/#organization",
      "name": "onpremisai.be",
      "url": "https://onpremisai.be/",
      "logo": "https://onpremisai.be/logo/opa-header.png",
      "description": "Private AI servers and local AI clusters for LLM cost control, token burning reduction and data privacy AI."
    },
    {
      "@type": "WebSite",
      "@id": "https://onpremisai.be/#website",
      "name": "onpremisai.be",
      "url": "https://onpremisai.be/",
      "inLanguage": [
        "en",
        "fr"
      ],
      "publisher": {
        "@id": "https://onpremisai.be/#organization"
      },
      "keywords": [
        "token burning",
        "LLM cost",
        "AI cost",
        "data privacy AI",
        "local cluster",
        "AI cluster",
        "private AI server",
        "private LLM",
        "on-premise AI",
        "GPU inference",
        "private RAG",
        "local inference",
        "enterprise AI infrastructure",
        "sovereign AI",
        "token cost optimization",
        "cloud AI alternative",
        "AI server pricing",
        "LLM cost reduction",
        "AI cost control",
        "local GPU cluster",
        "private chatbot",
        "document AI privacy",
        "embeddings cost",
        "inference cost",
        "agentic workflow cost",
        "coûts token",
        "combustion de tokens",
        "coût LLM",
        "coût IA",
        "confidentialité des données IA",
        "cluster local",
        "cluster IA",
        "serveur IA privé",
        "LLM privé",
        "IA sur site",
        "inférence GPU",
        "RAG privé",
        "inférence locale",
        "infrastructure IA d’entreprise",
        "IA souveraine",
        "optimisation des coûts token",
        "alternative à l’IA cloud",
        "tarification serveur IA",
        "réduction coût LLM",
        "contrôle des coûts IA",
        "cluster GPU local",
        "chatbot privé",
        "confidentialité documentaire IA",
        "coût des embeddings",
        "coût d’inférence",
        "coût des workflows agentiques"
      ]
    },
    {
      "@type": "Service",
      "@id": "https://onpremisai.be/#private-ai-cluster-service",
      "name": "Private AI server and local AI cluster integration",
      "alternateName": [
        "Serveur IA privé",
        "Cluster IA local",
        "On-premise AI infrastructure"
      ],
      "serviceType": "Private AI infrastructure, local AI cluster deployment, LLM cost optimization, data privacy AI",
      "provider": {
        "@id": "https://onpremisai.be/#organization"
      },
      "areaServed": "Worldwide",
      "description": "Ready-to-use private AI servers and local clusters that reduce token burning, control AI cost and keep enterprise data private."
    },
    {
      "@type": "FAQPage",
      "@id": "https://onpremisai.be/#faq",
      "mainEntity": [
        {
          "@type": "Question",
          "name": "How can a local AI cluster reduce token burning?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "A local AI cluster moves repeated inference, embeddings, RAG and agent workloads from pay-per-token cloud APIs to owned GPU capacity, making LLM cost more predictable."
          }
        },
        {
          "@type": "Question",
          "name": "Pourquoi un cluster IA local améliore-t-il la confidentialité des données IA ?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "Les prompts, documents, embeddings et réponses restent dans le réseau de l’entreprise avec des règles d’accès maîtrisées et une inférence locale."
          }
        },
        {
          "@type": "Question",
          "name": "What workloads fit a private AI server?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "Private chatbots, document search, RAG, coding assistants, support agents, workflow automation, local inference, private LLMs and secure enterprise AI workloads."
          }
        }
      ]
    },
    {
      "@type": "CollectionPage",
      "@id": "https://onpremisai.be/#model-catalog",
      "name": "AI model catalog for private deployments",
      "url": "https://onpremisai.be/models.html",
      "inLanguage": [
        "en",
        "fr"
      ],
      "about": [
        {
          "@type": "Thing",
          "name": "open-weight models"
        },
        {
          "@type": "Thing",
          "name": "private LLM deployment"
        },
        {
          "@type": "Thing",
          "name": "embedding models"
        },
        {
          "@type": "Thing",
          "name": "coding models"
        }
      ],
      "isPartOf": {
        "@id": "https://onpremisai.be/#website"
      }
    }
  ]
}