video1.0<iframe src="https://www.loom.com/embed/d1270097c6c148148bdbaf9a2f7bfe7e" frameborder="0" width="1532" height="1149" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>11491532Loomhttps://www.loom.com11491532https://cdn.loom.com/sessions/thumbnails/d1270097c6c148148bdbaf9a2f7bfe7e-ce345d4365ff587d.gif195.8376In this video, I explain the different interaction modalities with Kodi and how to choose the right models for auto-complete. Larger models are used in chat and commands, while smaller, faster models are optimized for auto-complete. I provide examples to help you decide which modality suits your needs best. Remember, auto-complete only considers local context due to latency demands. No remote context is included. Pay attention to when auto-complete makes suggestions and how to accept or remove them.