{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/c74f131b917145cda59ee136f3850a1d\" frameborder=\"0\" width=\"1920\" height=\"1440\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1440,"width":1920,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1440,"thumbnail_width":1920,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/c74f131b917145cda59ee136f3850a1d-ccc8dfe5d1246278.gif","duration":99.605333,"title":"Distributed Quantum Model Across Consumer Devices","description":"This Loom explains a distributed approach for running a Quenn model across consumer and server devices using Prima.cpp integrated with QM and gBrain. The team focuses on privacy and data ownership as the top concerns for AI adoption, and on enabling families to use existing cheap devices rather than buying new hardware. They build on an ICLR 2026 paper and deployed the model in a ring of multiple devices, assigning eight layers to a MacBook Air and 20 to a $10,000 Linux server paired together to produce responses. The demo shows rank 0 and rank 1 machines in tmux and similar performance for QM as well, with next steps including iOS builds and demos on iPad and iPhone plus extended work for more realistic mesh situations."}