{"slug":"seatable-runs-a-self-hosted-llm-on-a-hetzner-gpu-server-with-vllm-to-power-seatable-6-0-ai-automations","url":"https://findausecase.com/use-cases/seatable-runs-a-self-hosted-llm-on-a-hetzner-gpu-server-with-vllm-to-power-seatable-6-0-ai-automations","title":"SeaTable runs a self-hosted LLM on a Hetzner GPU server with vLLM to power SeaTable 6.0 AI automations","description":"SeaTable integrates AI-powered automations in SeaTable 6.0 by running a self-hosted large language model on a GPU server at Hetzner using the vLLM framework in a production SaaS environment, keeping control over data. Described in a CS3 2026 conference talk abstract.","company":"SeaTable","industry":"Technology & Software","aiCapabilities":["Generative AI"],"technology":["vLLM","Hetzner"],"deployment":"Unknown","solutionApproach":"SeaTable 6.0 integrates AI-powered automations by running a self-hosted large language model on a GPU server at Hetzner using the vLLM framework, operated in a production SaaS environment.","businessValue":"Running the model self-hosted lets SeaTable keep full control over data and customization. The talk abstract states no figures.","evidence":{"band":"high"},"sourceUrl":"https://indico.cern.ch/event/1560960/contributions/6903796/","dates":{"publishedAt":"2026-10-24T05:49:31.774Z","publishedAtSource":"pipeline","updatedAt":"2026-10-24T05:49:31.774Z"},"license":"Open for reading and citing with a link to https://findausecase.com/use-cases/seatable-runs-a-self-hosted-llm-on-a-hetzner-gpu-server-with-vllm-to-power-seatable-6-0-ai-automations. Bulk republication requires permission."}